source: branches/HeuristicLab.BottomUpTreeDistance/HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/Analyzers/SymbolicDataAnalysisInternalDiversityAnalyzer.cs @ 11486

#region License Information
/* HeuristicLab
 * Copyright (C) 2002-2014 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
 *
 * This file is part of HeuristicLab.
 *
 * HeuristicLab is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 *
 * HeuristicLab is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
 */
#endregion

using System;
using System.Linq;
using HeuristicLab.Analysis;
using HeuristicLab.Common;
using HeuristicLab.Core;
using HeuristicLab.Data;
using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
using HeuristicLab.Operators;
using HeuristicLab.Optimization;
using HeuristicLab.Parameters;
using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;

namespace HeuristicLab.Problems.DataAnalysis.Symbolic.Analyzers {
  [StorableClass]
  [Item("SymbolicDataAnalysisInternalDiversityAnalyzer", "An analyzer that determines the average diversity of symbolic expression trees.")]
  public class SymbolicDataAnalysisInternalDiversityAnalyzer : SingleSuccessorOperator, ISymbolicDataAnalysisAnalyzer {
    private const string SymbolicExpressionTreeParameterName = "SymbolicExpressionTree";
    private const string QualityParameterName = "Quality";
    private const string PercentageOfBestSolutionsParameterName = "PercentageOfBestSolutions";
    private const string ResultCollectionParameterName = "Results";
    private readonly BottomUpSimilarityCalculator busCalculator;

    public SymbolicDataAnalysisInternalDiversityAnalyzer() {
      busCalculator = new BottomUpSimilarityCalculator();

      Parameters.Add(new ScopeTreeLookupParameter<ISymbolicExpressionTree>(SymbolicExpressionTreeParameterName));
      Parameters.Add(new ScopeTreeLookupParameter<DoubleValue>(QualityParameterName));
      Parameters.Add(new LookupParameter<ResultCollection>(ResultCollectionParameterName));
      Parameters.Add(new FixedValueParameter<PercentValue>(PercentageOfBestSolutionsParameterName));
    }

    protected SymbolicDataAnalysisInternalDiversityAnalyzer(SymbolicDataAnalysisInternalDiversityAnalyzer original, Cloner cloner)
      : base(original, cloner) {
    }

    public override IDeepCloneable Clone(Cloner cloner) {
      return new SymbolicDataAnalysisInternalDiversityAnalyzer(this, cloner);
    }

    #region parameter properties
    public IScopeTreeLookupParameter<ISymbolicExpressionTree> SymbolicExpressionTreeParameter {
      get { return (IScopeTreeLookupParameter<ISymbolicExpressionTree>)Parameters[SymbolicExpressionTreeParameterName]; }
    }

    public IScopeTreeLookupParameter<DoubleValue> QualityParameter {
      get { return (IScopeTreeLookupParameter<DoubleValue>)Parameters[QualityParameterName]; }
    }

    public ILookupParameter<ResultCollection> ResultCollectionParameter {
      get { return (ILookupParameter<ResultCollection>)Parameters[ResultCollectionParameterName]; }
    }

    public IFixedValueParameter<PercentValue> PercentageOfBestSolutionsParameter {
      get { return (IFixedValueParameter<PercentValue>)Parameters[PercentageOfBestSolutionsParameterName]; }
    }
    #endregion

    public double PercentageOfBestSolutions {
      get { return PercentageOfBestSolutionsParameter.Value.Value; }
      set { PercentageOfBestSolutionsParameter.Value.Value = value; }
    }

    public bool EnabledByDefault { get { return true; } }

    public override IOperation Apply() {

      var results = ResultCollectionParameter.ActualValue;

      DataTable table;
      if (!results.ContainsKey("Avg. Internal Diversity")) {
        table = new DataTable("Internal Diversity") { VisualProperties = { YAxisTitle = "Internal Diversity" } };
        var row = new DataRow("Avg. Internal Diversity") { VisualProperties = { StartIndexZero = true } };
        table.Rows.Add(row);
        results.Add(new Result("Avg. Internal Diversity", table));
      } else {
        table = (DataTable)results["Avg. Internal Diversity"].Value;
      }

      var trees = SymbolicExpressionTreeParameter.ActualValue.ToArray();
      var qualities = QualityParameter.ActualValue.ToArray();

      var pairs = trees.Zip(qualities, (t, q) => new { Tree = t, Quality = q }).OrderByDescending(x => x.Quality);
      int n = (int)Math.Floor(trees.Length * PercentageOfBestSolutions);
      double avgInternalDiversity = pairs.Take(n).Average(x => CalculateInternalDiversity(x.Tree));

      table.Rows["Avg. Internal Diversity"].Values.Add(avgInternalDiversity);
      return base.Apply();
    }

    private double CalculateInternalDiversity(ISymbolicExpressionTree tree) {
      var branchPoint = tree.IterateNodesPrefix().FirstOrDefault(x => x.SubtreeCount == 2);
      if (branchPoint == null)
        return 1;

      var s1 = branchPoint.GetSubtree(0);
      var s2 = branchPoint.GetSubtree(1);

      // set parents to null because otherwise the bottom-up calculator will throw an exception
      var p1 = s1.Parent;
      s1.Parent = null;
      var p2 = s2.Parent;
      s2.Parent = null;

      var m = busCalculator.ComputeBottomUpMapping(s1, s2);
      var diversity = 1 - 2.0 * m.Count / (s1.GetLength() + s2.GetLength());

      // restore parents
      s1.Parent = p1;
      s2.Parent = p2;

      return diversity;
    }
  }
}