Free cookie consent management tool by TermsFeed Policy Generator

source: branches/DataAnalysis/HeuristicLab.Problems.DataAnalysis.MultiVariate.TimeSeriesPrognosis/3.3/Symbolic/SymbolicTimeSeriesPrognosisSolution.cs @ 4475

Last change on this file since 4475 was 4475, checked in by gkronber, 13 years ago

Fixed bugs in time series prognosis classes #1142.

File size: 8.0 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2010 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.Collections.Generic;
23using HeuristicLab.Common;
24using HeuristicLab.Core;
25using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
26using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding.Symbols;
27using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
28using HeuristicLab.Problems.DataAnalysis.MultiVariate.TimeSeriesPrognosis.Symbolic.Symbols;
29using HeuristicLab.Problems.DataAnalysis.Symbolic.Symbols;
30using System.Linq;
31using System.Drawing;
32using System;
33using HeuristicLab.Data;
34
35namespace HeuristicLab.Problems.DataAnalysis.MultiVariate.TimeSeriesPrognosis.Symbolic {
36  [StorableClass]
37  [Item("SymbolicTimeSeriesPrognosisSolution", "Represents a solution for time series prognosis.")]
38  public class SymbolicTimeSeriesPrognosisSolution : NamedItem, IMultiVariateDataAnalysisSolution, IStorableContent {
39    [Storable]
40    private MultiVariateDataAnalysisProblemData problemData;
41    [Storable]
42    private SymbolicTimeSeriesPrognosisModel model;
43    [Storable]
44    private int horizon;
45    [Storable]
46    private string conditionalEvaluationVariable;
47    [Storable]
48    private double[] lowerEstimationLimit;
49    [Storable]
50    private double[] upperEstimationLimit;
51
52    public string Filename { get; set; }
53
54    [StorableConstructor]
55    protected SymbolicTimeSeriesPrognosisSolution(bool deserializing) : base(deserializing) { }
56
57    public SymbolicTimeSeriesPrognosisSolution() {
58      horizon = 1;
59    }
60
61    public SymbolicTimeSeriesPrognosisSolution(MultiVariateDataAnalysisProblemData problemData, SymbolicTimeSeriesPrognosisModel model, int horizon, string conditionalEvaluationVariable, double[] lowerEstimationLimit, double[] upperEstimationLimit)
62      : this() {
63      this.problemData = problemData;
64      this.model = model;
65      this.horizon = horizon;
66      this.conditionalEvaluationVariable = conditionalEvaluationVariable;
67      this.lowerEstimationLimit = (double[])lowerEstimationLimit.Clone();
68      this.upperEstimationLimit = (double[])upperEstimationLimit.Clone();
69    }
70
71    [StorableHook(HookType.AfterDeserialization)]
72    private void Initialize() {
73      if (problemData != null)
74        RegisterProblemDataEvents();
75    }
76
77    public override Image ItemImage {
78      get { return HeuristicLab.Common.Resources.VS2008ImageLibrary.Function; }
79    }
80
81    public int Horizon {
82      get { return horizon; }
83      set {
84        if (value <= 0) throw new ArgumentException();
85        horizon = value;
86      }
87    }
88
89    public SymbolicTimeSeriesPrognosisModel Model {
90      get { return model; }
91      set {
92        if (model != value) {
93          if (value == null) throw new ArgumentNullException();
94          model = value;
95          RaiseModelChanged();
96        }
97      }
98    }
99
100    public string ConditionalEvaluationVariable {
101      get { return conditionalEvaluationVariable; }
102      set {
103        if (conditionalEvaluationVariable != value) {
104          conditionalEvaluationVariable = value;
105          RaiseEstimatedValuesChanged();
106        }
107      }
108    }
109
110    public double GetLowerEstimationLimit(int i) {
111      return lowerEstimationLimit[i];
112    }
113    public double GetUpperEstimationLimit(int i) {
114      return upperEstimationLimit[i];
115    }
116
117
118    public IEnumerable<double[]> GetPrognosis(int t) {
119      return model.GetEstimatedValues(problemData, t, t + 1);
120    }
121
122    #region IMultiVariateDataAnalysisSolution Members
123
124    public MultiVariateDataAnalysisProblemData ProblemData {
125      get { return problemData; }
126      set {
127        if (problemData != value) {
128          if (value == null) throw new ArgumentNullException();
129          if (model != null && problemData != null &&
130            !(problemData.InputVariables
131              .Select(c => c.Value)
132              .SequenceEqual(value.InputVariables
133                             .Select(c => c.Value)) &&
134             problemData.TargetVariables
135             .Select(c => c.Value)
136             .SequenceEqual(value.TargetVariables
137                            .Select(c => c.Value)))) {
138            throw new ArgumentException("Could not set new problem data with different structure");
139          }
140
141          if (problemData != null) DeregisterProblemDataEvents();
142          problemData = value;
143          RaiseProblemDataChanged();
144          RegisterProblemDataEvents();
145        }
146      }
147    }
148
149    IMultiVariateDataAnalysisModel IMultiVariateDataAnalysisSolution.Model {
150      get { return model; }
151    }
152
153    public IEnumerable<double[]> EstimatedValues {
154      get {
155        return ApplyEstimationLimits(model.GetEstimatedValues(problemData, 0, problemData.Dataset.Rows));
156      }
157    }
158
159    public IEnumerable<double[]> EstimatedTrainingValues {
160      get {
161        return ApplyEstimationLimits(model.GetEstimatedValues(problemData, problemData.TrainingSamplesStart.Value, problemData.TrainingSamplesEnd.Value));
162      }
163    }
164
165    public IEnumerable<double[]> EstimatedTestValues {
166      get {
167        return ApplyEstimationLimits(model.GetEstimatedValues(problemData, problemData.TestSamplesStart.Value, problemData.TestSamplesEnd.Value));
168      }
169    }
170
171    #endregion
172
173
174    #region Events
175    protected virtual void RegisterProblemDataEvents() {
176      ProblemData.ProblemDataChanged += new EventHandler(ProblemData_Changed);
177    }
178    protected virtual void DeregisterProblemDataEvents() {
179      ProblemData.ProblemDataChanged += new EventHandler(ProblemData_Changed);
180    }
181    private void ProblemData_Changed(object sender, EventArgs e) {
182      RaiseProblemDataChanged();
183    }
184
185    public event EventHandler ProblemDataChanged;
186    protected virtual void RaiseProblemDataChanged() {
187      var listeners = ProblemDataChanged;
188      if (listeners != null)
189        listeners(this, EventArgs.Empty);
190    }
191
192    public event EventHandler ModelChanged;
193    protected virtual void RaiseModelChanged() {
194      EventHandler handler = ModelChanged;
195      if (handler != null)
196        handler(this, EventArgs.Empty);
197    }
198
199    public event EventHandler EstimatedValuesChanged;
200    protected virtual void RaiseEstimatedValuesChanged() {
201      var listeners = EstimatedValuesChanged;
202      if (listeners != null)
203        listeners(this, EventArgs.Empty);
204    }
205    #endregion
206
207
208    public override IDeepCloneable Clone(Cloner cloner) {
209      SymbolicTimeSeriesPrognosisSolution clone = (SymbolicTimeSeriesPrognosisSolution)base.Clone(cloner);
210      clone.problemData = (MultiVariateDataAnalysisProblemData)cloner.Clone(problemData);
211      clone.model = (SymbolicTimeSeriesPrognosisModel)cloner.Clone(model);
212      clone.horizon = horizon;
213      clone.conditionalEvaluationVariable = conditionalEvaluationVariable;
214      clone.lowerEstimationLimit = (double[])lowerEstimationLimit.Clone();
215      clone.upperEstimationLimit = (double[])upperEstimationLimit.Clone();
216      return clone;
217    }
218
219    private IEnumerable<double[]> ApplyEstimationLimits(IEnumerable<double[]> values) {
220      foreach (var xs in values) {
221        for (int i = 0; i < xs.Length; i++) {
222          xs[i] = Math.Max(lowerEstimationLimit[i], Math.Min(upperEstimationLimit[i], xs[i]));
223        }
224        yield return xs;
225      }
226    }
227  }
228}
Note: See TracBrowser for help on using the repository browser.