source: branches/HeuristicLab.OSGAEvaluator/HeuristicLab.OSGAEvaluator/SymbolicRegressionSingleObjectiveOSGAEvaluator.cs @ 14280

#region License Information
/* HeuristicLab
 * Copyright (C) 2002-2016 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.Collections.Generic;
using System.Linq;
using HeuristicLab.Common;
using HeuristicLab.Core;
using HeuristicLab.Data;
using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
using HeuristicLab.Optimization;
using HeuristicLab.Parameters;
using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;

namespace HeuristicLab.Problems.DataAnalysis.Symbolic.Regression {
  [Item("SymbolicRegressionSingleObjectiveOSGAEvaluator", "An evaluator which tries to predict when a child will not be able to fullfil offspring selection criteria, to save evaluation time.")]
  [StorableClass]
  public class SymbolicRegressionSingleObjectiveOsgaEvaluator : SymbolicRegressionSingleObjectiveEvaluator {
    private const string RelativeParentChildQualityThresholdParameterName = "RelativeParentChildQualityThreshold";
    private const string RelativeFitnessEvaluationIntervalSizeParameterName = "RelativeFitnessEvaluationIntervalSize";
    private const string ResultCollectionParameterName = "Results";
    private const string AggregateStatisticsParameterName = "AggregateStatistics";
    private const string ActualSelectionPressureParameterName = "SelectionPressure";
    private const string UseAdaptiveQualityThresholdParameterName = "UseAdaptiveQualityThreshold";
    private const string UseFixedEvaluationIntervalsParameterName = "UseFixedEvaluationIntervals";

    #region parameters
    public IFixedValueParameter<BoolValue> UseFixedEvaluationIntervalsParameter {
      get { return (IFixedValueParameter<BoolValue>)Parameters[UseFixedEvaluationIntervalsParameterName]; }
    }
    public IFixedValueParameter<BoolValue> UseAdaptiveQualityThresholdParameter {
      get { return (IFixedValueParameter<BoolValue>)Parameters[UseAdaptiveQualityThresholdParameterName]; }
    }
    public ILookupParameter<DoubleValue> ActualSelectionPressureParameter {
      get { return (ILookupParameter<DoubleValue>)Parameters[ActualSelectionPressureParameterName]; }
    }
    public ILookupParameter<ResultCollection> ResultCollectionParameter {
      get { return (ILookupParameter<ResultCollection>)Parameters[ResultCollectionParameterName]; }
    }
    public IValueParameter<BoolValue> AggregateStatisticsParameter {
      get { return (IValueParameter<BoolValue>)Parameters[AggregateStatisticsParameterName]; }
    }
    public IValueParameter<IntMatrix> RejectedStatsParameter {
      get { return (IValueParameter<IntMatrix>)Parameters["RejectedStats"]; }
    }
    public IValueParameter<IntMatrix> NotRejectedStatsParameter {
      get { return (IValueParameter<IntMatrix>)Parameters["TotalStats"]; }
    }
    public IValueLookupParameter<DoubleValue> ComparisonFactorParameter {
      get { return (ValueLookupParameter<DoubleValue>)Parameters["ComparisonFactor"]; }
    }
    public IFixedValueParameter<PercentValue> RelativeParentChildQualityThresholdParameter {
      get { return (IFixedValueParameter<PercentValue>)Parameters[RelativeParentChildQualityThresholdParameterName]; }
    }
    public IFixedValueParameter<PercentValue> RelativeFitnessEvaluationIntervalSizeParameter {
      get { return (IFixedValueParameter<PercentValue>)Parameters[RelativeFitnessEvaluationIntervalSizeParameterName]; }
    }
    public IScopeTreeLookupParameter<DoubleValue> ParentQualitiesParameter { get { return (IScopeTreeLookupParameter<DoubleValue>)Parameters["ParentQualities"]; } }
    #endregion

    #region parameter properties
    public bool UseFixedEvaluationIntervals {
      get { return UseFixedEvaluationIntervalsParameter.Value.Value; }
      set { UseFixedEvaluationIntervalsParameter.Value.Value = value; }
    }
    public bool UseAdaptiveQualityThreshold {
      get { return UseAdaptiveQualityThresholdParameter.Value.Value; }
      set { UseAdaptiveQualityThresholdParameter.Value.Value = value; }
    }
    public double RelativeParentChildQualityThreshold {
      get { return RelativeParentChildQualityThresholdParameter.Value.Value; }
      set { RelativeParentChildQualityThresholdParameter.Value.Value = value; }
    }

    public double RelativeFitnessEvaluationIntervalSize {
      get { return RelativeFitnessEvaluationIntervalSizeParameter.Value.Value; }
      set { RelativeFitnessEvaluationIntervalSizeParameter.Value.Value = value; }
    }

    public IntMatrix RejectedStats {
      get { return RejectedStatsParameter.Value; }
      set { RejectedStatsParameter.Value = value; }
    }

    public IntMatrix TotalStats {
      get { return NotRejectedStatsParameter.Value; }
      set { NotRejectedStatsParameter.Value = value; }
    }
    #endregion

    public override bool Maximization {
      get { return true; }
    }

    public SymbolicRegressionSingleObjectiveOsgaEvaluator() {
      Parameters.Add(new ValueLookupParameter<DoubleValue>("ComparisonFactor", "Determines if the quality should be compared to the better parent (1.0), to the worse (0.0) or to any linearly interpolated value between them."));
      Parameters.Add(new FixedValueParameter<PercentValue>(RelativeParentChildQualityThresholdParameterName, new PercentValue(0.9)));
      Parameters.Add(new FixedValueParameter<PercentValue>(RelativeFitnessEvaluationIntervalSizeParameterName, new PercentValue(0.1)));
      Parameters.Add(new LookupParameter<ResultCollection>(ResultCollectionParameterName));
      Parameters.Add(new ScopeTreeLookupParameter<DoubleValue>("ParentQualities") { ActualName = "Quality" });
      Parameters.Add(new ValueParameter<IntMatrix>("RejectedStats", new IntMatrix()));
      Parameters.Add(new ValueParameter<IntMatrix>("TotalStats", new IntMatrix()));
      Parameters.Add(new ValueParameter<BoolValue>(AggregateStatisticsParameterName, new BoolValue(false)));
      Parameters.Add(new LookupParameter<DoubleValue>(ActualSelectionPressureParameterName));
      Parameters.Add(new FixedValueParameter<BoolValue>(UseAdaptiveQualityThresholdParameterName, new BoolValue(false)));
      Parameters.Add(new FixedValueParameter<BoolValue>(UseFixedEvaluationIntervalsParameterName, new BoolValue(false)));
    }

    [StorableHook(HookType.AfterDeserialization)]
    private void AfterDeserialization() {
      if (!Parameters.ContainsKey(ActualSelectionPressureParameterName))
        Parameters.Add(new LookupParameter<DoubleValue>(ActualSelectionPressureParameterName));

      if (!Parameters.ContainsKey(UseAdaptiveQualityThresholdParameterName))
        Parameters.Add(new FixedValueParameter<BoolValue>(UseAdaptiveQualityThresholdParameterName, new BoolValue(false)));

      if (!Parameters.ContainsKey(UseFixedEvaluationIntervalsParameterName))
        Parameters.Add(new FixedValueParameter<BoolValue>(UseFixedEvaluationIntervalsParameterName, new BoolValue(false)));
    }

    [StorableConstructor]
    protected SymbolicRegressionSingleObjectiveOsgaEvaluator(bool deserializing) : base(deserializing) { }

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

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

    public override void ClearState() {
      base.ClearState();
      RejectedStats = new IntMatrix();
      TotalStats = new IntMatrix();
    }

    public override IOperation InstrumentedApply() {
      var solution = SymbolicExpressionTreeParameter.ActualValue;
      IEnumerable<int> rows = GenerateRowsToEvaluate();

      var interpreter = SymbolicDataAnalysisTreeInterpreterParameter.ActualValue;
      var estimationLimits = EstimationLimitsParameter.ActualValue;
      var problemData = ProblemDataParameter.ActualValue;
      var applyLinearScaling = ApplyLinearScalingParameter.ActualValue.Value;

      double quality;
      var parentQualities = ParentQualitiesParameter.ActualValue;

      // parent subscopes are not present during evaluation of the initial population
      if (parentQualities.Length > 0) {
        quality = Calculate(interpreter, solution, estimationLimits, problemData, rows);
      } else {
        quality = Calculate(interpreter, solution, estimationLimits.Lower, estimationLimits.Upper, problemData, rows, applyLinearScaling);
      }
      QualityParameter.ActualValue = new DoubleValue(quality);

      return base.InstrumentedApply();
    }

    public static double Calculate(ISymbolicDataAnalysisExpressionTreeInterpreter interpreter, ISymbolicExpressionTree solution, double lowerEstimationLimit, double upperEstimationLimit, IRegressionProblemData problemData, IEnumerable<int> rows, bool applyLinearScaling) {
      IEnumerable<double> estimatedValues = interpreter.GetSymbolicExpressionTreeValues(solution, problemData.Dataset, rows);
      IEnumerable<double> targetValues = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, rows);
      OnlineCalculatorError errorState;

      double r;
      if (applyLinearScaling) {
        var rCalculator = new OnlinePearsonsRCalculator();
        CalculateWithScaling(targetValues, estimatedValues, lowerEstimationLimit, upperEstimationLimit, rCalculator, problemData.Dataset.Rows);
        errorState = rCalculator.ErrorState;
        r = rCalculator.R;
      } else {
        IEnumerable<double> boundedEstimatedValues = estimatedValues.LimitToRange(lowerEstimationLimit, upperEstimationLimit);
        r = OnlinePearsonsRCalculator.Calculate(targetValues, boundedEstimatedValues, out errorState);
      }
      if (errorState != OnlineCalculatorError.None) return double.NaN;
      return r * r;
    }

    private double Calculate(ISymbolicDataAnalysisExpressionTreeInterpreter interpreter, ISymbolicExpressionTree solution, DoubleLimit estimationLimits, IRegressionProblemData problemData, IEnumerable<int> rows) {
      var estimatedValues = interpreter.GetSymbolicExpressionTreeValues(solution, problemData.Dataset, rows).LimitToRange(estimationLimits.Lower, estimationLimits.Upper);
      var targetValues = problemData.Dataset.GetReadOnlyDoubleValues(problemData.TargetVariable);

      var parentQualities = ParentQualitiesParameter.ActualValue.Select(x => x.Value);
      var minQuality = parentQualities.Min();
      var maxQuality = parentQualities.Max();
      var comparisonFactor = ComparisonFactorParameter.ActualValue.Value;
      var parentQuality = minQuality + (maxQuality - minQuality) * comparisonFactor;

      var e = estimatedValues.GetEnumerator();

      #region fixed intervals
      if (UseFixedEvaluationIntervals) {
        double threshold = parentQuality * RelativeParentChildQualityThreshold;
        if (UseAdaptiveQualityThreshold) {
          var actualSelectionPressure = ActualSelectionPressureParameter.ActualValue;
          if (actualSelectionPressure != null)
            threshold = parentQuality * (1 - actualSelectionPressure.Value / 100.0);
        }

        var pearsonRCalculator = new OnlinePearsonsRCalculator();
        var targetValuesEnumerator = targetValues.GetEnumerator();
        var trainingPartitionSize = problemData.TrainingPartition.Size;
        var interval = (int)Math.Floor(trainingPartitionSize * RelativeFitnessEvaluationIntervalSize);

        var aggregateStatistics = AggregateStatisticsParameter.Value.Value;
        var i = 0;
        if (aggregateStatistics) {
          var trainingEnd = problemData.TrainingPartition.End;
          var qualityPerInterval = new List<double>();
          while (targetValuesEnumerator.MoveNext() && e.MoveNext()) {
            pearsonRCalculator.Add(targetValuesEnumerator.Current, e.Current);
            ++i;
            if (i % interval == 0 || i == trainingPartitionSize) {
              var q = pearsonRCalculator.ErrorState != OnlineCalculatorError.None ? double.NaN : pearsonRCalculator.R;
              qualityPerInterval.Add(q * q);
            }
          }
          var r = pearsonRCalculator.ErrorState != OnlineCalculatorError.None ? double.NaN : pearsonRCalculator.R;
          var actualQuality = r * r;

          bool predictedRejected = false;

          i = 0;
          double quality = actualQuality;
          foreach (var q in qualityPerInterval) {
            if (double.IsNaN(q) || !(q > threshold)) {
              predictedRejected = true;
              quality = q;
              break;
            }
            ++i;
          }

          var actuallyRejected = !(actualQuality > parentQuality);

          if (RejectedStats.Rows == 0 || TotalStats.Rows == 0) {
            RejectedStats = new IntMatrix(2, qualityPerInterval.Count);
            RejectedStats.RowNames = new[] { "Predicted", "Actual" };
            RejectedStats.ColumnNames = Enumerable.Range(1, RejectedStats.Columns).Select(x => string.Format("0-{0}", Math.Min(trainingEnd, x * interval)));
            TotalStats = new IntMatrix(2, 2);
            TotalStats.RowNames = new[] { "Predicted", "Actual" };
            TotalStats.ColumnNames = new[] { "Rejected", "Not Rejected" };
          }
          // gather some statistics
          if (predictedRejected) {
            RejectedStats[0, i]++;
            TotalStats[0, 0]++;
          } else {
            TotalStats[0, 1]++;
          }
          if (actuallyRejected) {
            TotalStats[1, 0]++;
          } else {
            TotalStats[1, 1]++;
          }
          if (predictedRejected && actuallyRejected) {
            RejectedStats[1, i]++;
          }
          return quality;
        } else {
          while (targetValuesEnumerator.MoveNext() && e.MoveNext()) {
            pearsonRCalculator.Add(targetValuesEnumerator.Current, e.Current);
            ++i;
            if (i % interval == 0 || i == trainingPartitionSize) {
              var q = pearsonRCalculator.ErrorState != OnlineCalculatorError.None ? double.NaN : pearsonRCalculator.R;
              var quality = q * q;
              if (!(quality > threshold))
                return quality;
            }
          }
          var r = pearsonRCalculator.ErrorState != OnlineCalculatorError.None ? double.NaN : pearsonRCalculator.R;
          var actualQuality = r * r;
          return actualQuality;
        }
        #endregion
      } else {
        var calculator = new OnlinePearsonsRCalculator();
        var trainingPartitionSize = problemData.TrainingPartition.Size;
        var interval = (int)Math.Floor(trainingPartitionSize * RelativeFitnessEvaluationIntervalSize);
        double quality = double.NaN;
        var estimated = new List<double>(); // save estimated values in a list so we don't re-evaluate 
        // use the actual estimated values for the first i * interval rows of the training partition and and assume the remaining rows are perfectly correlated
        // if the quality of the individual still falls below the parent quality, then we can reject it sooner, otherwise as i increases the whole estimated series will be used
        for (int i = 0; i < trainingPartitionSize; i += interval) {
          calculator.Reset();
          // save estimated values into the list (for caching)
          int j = i;
          int end = Math.Min(trainingPartitionSize, i + interval);
          while (j < end && e.MoveNext()) {
            estimated.Add(e.Current);
            j++;
          }
          var start = problemData.TrainingPartition.Start;
          // add (estimated, target) pairs to the calculator
          for (j = 0; j < end; ++j) {
            var index = j + start;
            calculator.Add(targetValues[index], estimated[j]);
          }
          // add (target, target) pairs to the calculator (simulate perfect correlation on the remaining rows)
          for (; j < trainingPartitionSize; ++j) {
            var index = j + start;
            var v = targetValues[index];
            calculator.Add(v, v);
          }
          var r = calculator.ErrorState == OnlineCalculatorError.None ? calculator.R : double.NaN;
          quality = r * r;
          if (!(quality > parentQuality))
            break;
        }
        return quality;
      }
    }

    public override double Evaluate(IExecutionContext context, ISymbolicExpressionTree tree, IRegressionProblemData problemData, IEnumerable<int> rows) {
      SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = context;
      EstimationLimitsParameter.ExecutionContext = context;
      ApplyLinearScalingParameter.ExecutionContext = context;

      var interpreter = SymbolicDataAnalysisTreeInterpreterParameter.ActualValue;
      var estimationLimits = EstimationLimitsParameter.ActualValue;
      var applyLinearScaling = ApplyLinearScalingParameter.ActualValue.Value;

      double r2 = Calculate(interpreter, tree, estimationLimits.Lower, estimationLimits.Upper, problemData, rows, applyLinearScaling);

      SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = null;
      EstimationLimitsParameter.ExecutionContext = null;
      ApplyLinearScalingParameter.ExecutionContext = null;

      return r2;
    }
  }
}