source: branches/DataAnalysis Refactoring/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SymbolicRegressionSolution.cs @ 5736

Last change on this file since 5736 was 5736, checked in by gkronber, 11 years ago

#1418 implemented linear scaling for classification solutions, fixed bugs interactive simplifier view for classification solutions.

File size: 6.4 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2011 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 System.Linq;
24using HeuristicLab.Common;
25using HeuristicLab.Core;
26using HeuristicLab.Data;
27using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
28using HeuristicLab.Operators;
29using HeuristicLab.Parameters;
30using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
31using HeuristicLab.Optimization;
32using System;
33
34namespace HeuristicLab.Problems.DataAnalysis.Symbolic.Regression {
35  /// <summary>
36  /// Represents a symbolic regression solution (model + data) and attributes of the solution like accuracy and complexity
37  /// </summary>
38  [StorableClass]
39  [Item(Name = "SymbolicRegressionSolution", Description = "Represents a symbolic regression solution (model + data) and attributes of the solution like accuracy and complexity.")]
40  public sealed class SymbolicRegressionSolution : RegressionSolution, ISymbolicRegressionSolution {
41    private const string ModelLengthResultName = "ModelLength";
42    private const string ModelDepthResultName = "ModelDepth";
43
44    public new ISymbolicRegressionModel Model {
45      get { return (ISymbolicRegressionModel)base.Model; }
46      set { base.Model = value; }
47    }
48    ISymbolicDataAnalysisModel ISymbolicDataAnalysisSolution.Model {
49      get { return (ISymbolicDataAnalysisModel)base.Model; }
50    }
51    public int ModelLength {
52      get { return ((IntValue)this[ModelLengthResultName].Value).Value; }
53      private set { ((IntValue)this[ModelLengthResultName].Value).Value = value; }
54    }
55
56    public int ModelDepth {
57      get { return ((IntValue)this[ModelDepthResultName].Value).Value; }
58      private set { ((IntValue)this[ModelDepthResultName].Value).Value = value; }
59    }
60
61    [StorableConstructor]
62    private SymbolicRegressionSolution(bool deserializing) : base(deserializing) { }
63    private SymbolicRegressionSolution(SymbolicRegressionSolution original, Cloner cloner)
64      : base(original, cloner) {
65    }
66    public SymbolicRegressionSolution(ISymbolicRegressionModel model, IRegressionProblemData problemData)
67      : base(model, problemData) {
68      Add(new Result(ModelLengthResultName, "Length of the symbolic regression model.", new IntValue()));
69      Add(new Result(ModelDepthResultName, "Depth of the symbolic regression model.", new IntValue()));
70      RecalculateResults();
71    }
72
73    public override IDeepCloneable Clone(Cloner cloner) {
74      return new SymbolicRegressionSolution(this, cloner);
75    }
76
77    protected override void OnModelChanged(EventArgs e) {
78      base.OnModelChanged(e);
79      RecalculateResults();
80    }
81
82    private new void RecalculateResults() {
83      ModelLength = Model.SymbolicExpressionTree.Length;
84      ModelDepth = Model.SymbolicExpressionTree.Depth;
85    }
86
87    public void ScaleModel() {
88      var dataset = ProblemData.Dataset;
89      var targetVariable = ProblemData.TargetVariable;
90      var rows = ProblemData.TrainingIndizes;
91      var estimatedValues = GetEstimatedValues(rows);
92      var targetValues = dataset.GetEnumeratedVariableValues(targetVariable, rows);
93      double alpha;
94      double beta;
95      OnlineLinearScalingParameterCalculator.Calculate(estimatedValues, targetValues, out alpha, out beta);
96
97      ConstantTreeNode alphaTreeNode = null;
98      ConstantTreeNode betaTreeNode = null;
99      // check if model has been scaled previously by analyzing the structure of the tree
100      var startNode = Model.SymbolicExpressionTree.Root.GetSubtree(0);
101      if (startNode.GetSubtree(0).Symbol is Addition) {
102        var addNode = startNode.GetSubtree(0);
103        if (addNode.SubtreesCount == 2 && addNode.GetSubtree(0).Symbol is Multiplication && addNode.GetSubtree(1).Symbol is Constant) {
104          alphaTreeNode = addNode.GetSubtree(1) as ConstantTreeNode;
105          var mulNode = addNode.GetSubtree(0);
106          if (mulNode.SubtreesCount == 2 && mulNode.GetSubtree(1).Symbol is Constant) {
107            betaTreeNode = mulNode.GetSubtree(1) as ConstantTreeNode;
108          }
109        }
110      }
111      // if tree structure matches the structure necessary for linear scaling then reuse the existing tree nodes
112      if (alphaTreeNode != null && betaTreeNode != null) {
113        betaTreeNode.Value *= beta;
114        alphaTreeNode.Value *= beta;
115        alphaTreeNode.Value += alpha;
116      } else {
117        var mainBranch = startNode.GetSubtree(0);
118        startNode.RemoveSubtree(0);
119        var scaledMainBranch = MakeSum(MakeProduct(beta, mainBranch), alpha);
120        startNode.AddSubtree(scaledMainBranch);
121      }
122
123      OnModelChanged(EventArgs.Empty);
124    }
125
126    private static ISymbolicExpressionTreeNode MakeSum(ISymbolicExpressionTreeNode treeNode, double alpha) {
127      if (alpha.IsAlmost(0.0)) {
128        return treeNode;
129      } else {
130        var node = (new Addition()).CreateTreeNode();
131        var alphaConst = MakeConstant(alpha);
132        node.AddSubtree(treeNode);
133        node.AddSubtree(alphaConst);
134        return node;
135      }
136    }
137
138    private static ISymbolicExpressionTreeNode MakeProduct(double beta, ISymbolicExpressionTreeNode treeNode) {
139      if (beta.IsAlmost(1.0)) {
140        return treeNode;
141      } else {
142        var node = (new Multiplication()).CreateTreeNode();
143        var betaConst = MakeConstant(beta);
144        node.AddSubtree(treeNode);
145        node.AddSubtree(betaConst);
146        return node;
147      }
148    }
149
150    private static ISymbolicExpressionTreeNode MakeConstant(double c) {
151      var node = (ConstantTreeNode)(new Constant()).CreateTreeNode();
152      node.Value = c;
153      return node;
154    }
155  }
156}
Note: See TracBrowser for help on using the repository browser.