1 | #region License Information
|
---|
2 | /* HeuristicLab
|
---|
3 | * Copyright (C) 2002-2013 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 |
|
---|
22 | using System;
|
---|
23 | using System.Collections;
|
---|
24 | using System.Collections.Generic;
|
---|
25 | using System.Linq;
|
---|
26 | using HeuristicLab.Common;
|
---|
27 | using HeuristicLab.Core;
|
---|
28 | using HeuristicLab.Data;
|
---|
29 | using HeuristicLab.Problems.DataAnalysis;
|
---|
30 |
|
---|
31 | namespace HeuristicLab.DataPreprocessing {
|
---|
32 |
|
---|
33 | internal class PDSnapshot {
|
---|
34 | public IDictionary<int, IList> VariableValues { get; set; }
|
---|
35 |
|
---|
36 | public IList<string> VariableNames { get; set; }
|
---|
37 |
|
---|
38 | public double TrainingToTestRatio { get; set; }
|
---|
39 |
|
---|
40 | public DataPreprocessingChangedEventType ChangedType { get; set; }
|
---|
41 |
|
---|
42 | public int ChangedColumn { get; set; }
|
---|
43 |
|
---|
44 | public int ChangedRow { get; set; }
|
---|
45 | }
|
---|
46 |
|
---|
47 | [Item("PreprocessingData", "Represents data used for preprocessing.")]
|
---|
48 | public class PreprocessingData : NamedItem, IPreprocessingData {
|
---|
49 |
|
---|
50 | private const int MAX_UNDO_DEPTH = 5;
|
---|
51 |
|
---|
52 | private IDictionary<int, IList> variableValues;
|
---|
53 |
|
---|
54 | private IList<string> variableNames;
|
---|
55 |
|
---|
56 | private double trainingToTestRatio;
|
---|
57 |
|
---|
58 | private IList<PDSnapshot> undoHistory;
|
---|
59 |
|
---|
60 | //TODO: refactor extract Transaction logic in a own class
|
---|
61 | private int transactionDepth = 0;
|
---|
62 |
|
---|
63 | private PreprocessingData(PreprocessingData original, Cloner cloner)
|
---|
64 | : base(original, cloner) {
|
---|
65 | variableValues = CopyVariableValues(original.variableValues);
|
---|
66 | variableNames = new List<string>(original.variableNames);
|
---|
67 | trainingToTestRatio = original.trainingToTestRatio;
|
---|
68 | undoHistory = new List<PDSnapshot>();
|
---|
69 | }
|
---|
70 |
|
---|
71 | public PreprocessingData(IDataAnalysisProblemData problemData)
|
---|
72 | : base() {
|
---|
73 | Name = "-";
|
---|
74 |
|
---|
75 | variableNames = new List<string>(problemData.Dataset.VariableNames);
|
---|
76 | // create dictionary from variable name to index
|
---|
77 |
|
---|
78 | int columnIndex = 0;
|
---|
79 | variableValues = new Dictionary<int, IList>();
|
---|
80 | foreach (var variableName in problemData.Dataset.VariableNames) {
|
---|
81 | if (problemData.Dataset.IsType<double>(variableName)) {
|
---|
82 | variableValues[columnIndex] = problemData.Dataset.GetDoubleValues(variableName).ToList();
|
---|
83 | } else if (problemData.Dataset.IsType<string>(variableName)) {
|
---|
84 | variableValues[columnIndex] = CreateColumn<string>(problemData.Dataset, columnIndex, x => x);
|
---|
85 | } else if (problemData.Dataset.IsType<DateTime>(variableName)) {
|
---|
86 | variableValues[columnIndex] = CreateColumn<DateTime>(problemData.Dataset, columnIndex, x => DateTime.Parse(x));
|
---|
87 | } else {
|
---|
88 | throw new ArgumentException("The datatype of column " + variableName + " must be of type List<double>, List<string> or List<DateTime>");
|
---|
89 | }
|
---|
90 | ++columnIndex;
|
---|
91 | }
|
---|
92 |
|
---|
93 | trainingToTestRatio = (double)problemData.TrainingPartition.Size / Math.Max(problemData.Dataset.Rows, double.Epsilon);
|
---|
94 | undoHistory = new List<PDSnapshot>();
|
---|
95 | }
|
---|
96 |
|
---|
97 | private static IList CreateColumn<T>(Dataset ds, int column, Func<string, T> selector) {
|
---|
98 | var list = new List<T>(ds.Rows);
|
---|
99 | for (int row = 0; row < ds.Rows; ++row) {
|
---|
100 | list.Add(selector(ds.GetValue(row, column)));
|
---|
101 | }
|
---|
102 | return list;
|
---|
103 | }
|
---|
104 |
|
---|
105 | private IDictionary<int, IList> CopyVariableValues(IDictionary<int, IList> original) {
|
---|
106 | var copy = new Dictionary<int, IList>(variableValues);
|
---|
107 | for (int i = 0; i < original.Count; i++) {
|
---|
108 | variableValues[i] = (IList)Activator.CreateInstance(original[i].GetType(), original[i]);
|
---|
109 | }
|
---|
110 | return copy;
|
---|
111 | }
|
---|
112 |
|
---|
113 | private void SaveSnapshot(DataPreprocessingChangedEventType changedType, int column, int row) {
|
---|
114 | if (transactionDepth > 0) return;
|
---|
115 |
|
---|
116 | PDSnapshot currentSnapshot = new PDSnapshot();
|
---|
117 | currentSnapshot.VariableValues = CopyVariableValues(variableValues);
|
---|
118 | currentSnapshot.VariableNames = new List<string>(variableNames);
|
---|
119 | currentSnapshot.TrainingToTestRatio = trainingToTestRatio;
|
---|
120 | currentSnapshot.ChangedType = changedType;
|
---|
121 | currentSnapshot.ChangedColumn = column;
|
---|
122 | currentSnapshot.ChangedRow = row;
|
---|
123 |
|
---|
124 | if (undoHistory.Count >= MAX_UNDO_DEPTH)
|
---|
125 | undoHistory.RemoveAt(0);
|
---|
126 |
|
---|
127 | undoHistory.Add(currentSnapshot);
|
---|
128 | }
|
---|
129 |
|
---|
130 | #region NamedItem abstract Member Implementations
|
---|
131 |
|
---|
132 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
133 | return new PreprocessingData(this, cloner);
|
---|
134 | }
|
---|
135 |
|
---|
136 | #endregion
|
---|
137 |
|
---|
138 | #region IPreprocessingData Members
|
---|
139 |
|
---|
140 | public T GetCell<T>(int columnIndex, int rowIndex) {
|
---|
141 | return (T)variableValues[columnIndex][rowIndex];
|
---|
142 | }
|
---|
143 |
|
---|
144 |
|
---|
145 | public void SetCell<T>(int columnIndex, int rowIndex, T value) {
|
---|
146 | SaveSnapshot(DataPreprocessingChangedEventType.ChangeItem, columnIndex, rowIndex);
|
---|
147 | variableValues[columnIndex][rowIndex] = value;
|
---|
148 | if (transactionDepth <= 0)
|
---|
149 | OnChanged(DataPreprocessingChangedEventType.ChangeItem, columnIndex, rowIndex);
|
---|
150 | }
|
---|
151 |
|
---|
152 |
|
---|
153 | public string GetCellAsString(int columnIndex, int rowIndex) {
|
---|
154 | return variableValues[columnIndex][rowIndex].ToString();
|
---|
155 | }
|
---|
156 |
|
---|
157 |
|
---|
158 | [Obsolete("use the index based variant, is faster")]
|
---|
159 | public IList<T> GetValues<T>(string variableName) {
|
---|
160 | return GetValues<T>(GetColumnIndex(variableName));
|
---|
161 | }
|
---|
162 |
|
---|
163 | public IList<T> GetValues<T>(int columnIndex) {
|
---|
164 | return (IList<T>)variableValues[columnIndex];
|
---|
165 | }
|
---|
166 |
|
---|
167 | public void SetValues<T>(int columnIndex, IList<T> values) {
|
---|
168 | if (IsType<T>(columnIndex)) {
|
---|
169 | SaveSnapshot(DataPreprocessingChangedEventType.ChangeColumn, columnIndex, -1);
|
---|
170 | variableValues[columnIndex] = (IList)values;
|
---|
171 | } else {
|
---|
172 | throw new ArgumentException("The datatype of column " + columnIndex + " must be of type " + variableValues[columnIndex].GetType().Name + " but was " + typeof(T).Name);
|
---|
173 | }
|
---|
174 | if (transactionDepth <= 0)
|
---|
175 | OnChanged(DataPreprocessingChangedEventType.ChangeColumn, columnIndex, -1);
|
---|
176 | }
|
---|
177 |
|
---|
178 | public void InsertRow(int rowIndex) {
|
---|
179 | SaveSnapshot(DataPreprocessingChangedEventType.DeleteRow, -1, rowIndex);
|
---|
180 | foreach (IList column in variableValues.Values) {
|
---|
181 | Type type = column.GetType().GetGenericArguments()[0];
|
---|
182 | column.Insert(rowIndex, type.IsValueType ? Activator.CreateInstance(type) : null);
|
---|
183 | }
|
---|
184 | if (transactionDepth <= 0)
|
---|
185 | OnChanged(DataPreprocessingChangedEventType.AddRow, -1, rowIndex);
|
---|
186 | }
|
---|
187 |
|
---|
188 | public void DeleteRow(int rowIndex) {
|
---|
189 | SaveSnapshot(DataPreprocessingChangedEventType.AddRow, -1, rowIndex);
|
---|
190 | foreach (IList column in variableValues.Values) {
|
---|
191 | column.RemoveAt(rowIndex);
|
---|
192 | }
|
---|
193 | if (transactionDepth <= 0)
|
---|
194 | OnChanged(DataPreprocessingChangedEventType.DeleteRow, -1, rowIndex);
|
---|
195 | }
|
---|
196 |
|
---|
197 | public void InsertColumn<T>(string variableName, int columnIndex) {
|
---|
198 | SaveSnapshot(DataPreprocessingChangedEventType.DeleteColumn, columnIndex, -1);
|
---|
199 | variableValues.Add(columnIndex, new List<T>(Rows));
|
---|
200 | variableNames.Insert(columnIndex, variableName);
|
---|
201 | if (transactionDepth <= 0)
|
---|
202 | OnChanged(DataPreprocessingChangedEventType.AddColumn, columnIndex, -1);
|
---|
203 | }
|
---|
204 |
|
---|
205 | public void DeleteColumn(int columnIndex) {
|
---|
206 | SaveSnapshot(DataPreprocessingChangedEventType.AddColumn, columnIndex, -1);
|
---|
207 | variableValues.Remove(columnIndex);
|
---|
208 | variableNames.RemoveAt(columnIndex);
|
---|
209 | if (transactionDepth <= 0)
|
---|
210 | OnChanged(DataPreprocessingChangedEventType.DeleteColumn, columnIndex, -1);
|
---|
211 | }
|
---|
212 |
|
---|
213 | public IntRange TrainingPartition {
|
---|
214 | get { return new IntRange(0, (int)(Rows * trainingToTestRatio)); }
|
---|
215 | }
|
---|
216 |
|
---|
217 | public IntRange TestPartition {
|
---|
218 | get { return new IntRange((int)(Rows * trainingToTestRatio), Rows); }
|
---|
219 | }
|
---|
220 |
|
---|
221 | public string GetVariableName(int columnIndex) {
|
---|
222 | return variableNames[columnIndex];
|
---|
223 | }
|
---|
224 |
|
---|
225 | public IEnumerable<string> VariableNames {
|
---|
226 | get { return variableNames; }
|
---|
227 | }
|
---|
228 |
|
---|
229 | public int GetColumnIndex(string variableName) {
|
---|
230 | return variableNames.IndexOf(variableName);
|
---|
231 | }
|
---|
232 |
|
---|
233 | public bool IsType<T>(int columnIndex) {
|
---|
234 | return variableValues[columnIndex] is List<T>;
|
---|
235 | }
|
---|
236 |
|
---|
237 | public int Columns {
|
---|
238 | get { return variableNames.Count; }
|
---|
239 | }
|
---|
240 |
|
---|
241 | public int Rows {
|
---|
242 | get { return variableValues.Count > 0 ? variableValues[0].Count : 0; }
|
---|
243 | }
|
---|
244 |
|
---|
245 | public Dataset ExportToDataset() {
|
---|
246 | IList<IList> values = new List<IList>();
|
---|
247 |
|
---|
248 | for (int i = 0; i < Columns; ++i) {
|
---|
249 | values.Add(variableValues[i]);
|
---|
250 | }
|
---|
251 |
|
---|
252 | var dataset = new Dataset(variableNames, values);
|
---|
253 | return dataset;
|
---|
254 | }
|
---|
255 |
|
---|
256 | public event DataPreprocessingChangedEventHandler Changed;
|
---|
257 | protected virtual void OnChanged(DataPreprocessingChangedEventType type, int column, int row) {
|
---|
258 | var listeners = Changed;
|
---|
259 | if (listeners != null) listeners(this, new DataPreprocessingChangedEventArgs(type, column, row));
|
---|
260 | }
|
---|
261 |
|
---|
262 | public bool IsUndoAvailable {
|
---|
263 | get { return undoHistory.Count > 0; }
|
---|
264 | }
|
---|
265 |
|
---|
266 | public void Undo() {
|
---|
267 | if (IsUndoAvailable) {
|
---|
268 | PDSnapshot previousSnapshot = undoHistory[undoHistory.Count - 1];
|
---|
269 | variableValues = previousSnapshot.VariableValues;
|
---|
270 | variableNames = previousSnapshot.VariableNames;
|
---|
271 | trainingToTestRatio = previousSnapshot.TrainingToTestRatio;
|
---|
272 | undoHistory.Remove(previousSnapshot);
|
---|
273 | OnChanged(previousSnapshot.ChangedType,
|
---|
274 | previousSnapshot.ChangedColumn,
|
---|
275 | previousSnapshot.ChangedRow);
|
---|
276 | }
|
---|
277 | }
|
---|
278 |
|
---|
279 | public void BeginTransaction() {
|
---|
280 | SaveSnapshot(DataPreprocessingChangedEventType.Any, -1, -1);
|
---|
281 | transactionDepth++;
|
---|
282 | }
|
---|
283 |
|
---|
284 | public void EndTransaction() {
|
---|
285 | transactionDepth--;
|
---|
286 | if (transactionDepth < 0)
|
---|
287 | throw new InvalidOperationException("There is no open transaction that can be ended.");
|
---|
288 | if (transactionDepth == 0)
|
---|
289 | OnChanged(DataPreprocessingChangedEventType.Any, -1, -1);
|
---|
290 | }
|
---|
291 |
|
---|
292 | #endregion
|
---|
293 | }
|
---|
294 | }
|
---|