#region License Information
/* HeuristicLab
* Copyright (C) 2002-2010 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 .
*/
#endregion
using System;
using System.Collections.Generic;
using System.Linq;
using HeuristicLab.Core;
using HeuristicLab.Data;
using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
using HeuristicLab.Operators;
using HeuristicLab.Parameters;
using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
using HeuristicLab.Problems.DataAnalysis.Regression;
using HeuristicLab.Problems.DataAnalysis.Symbolic;
namespace HeuristicLab.Problems.DataAnalysis.MultiVariate.Regression.Symbolic.Evaluators {
[StorableClass]
public abstract class SymbolicVectorRegressionEvaluator : SingleSuccessorOperator, IMultiVariateDataAnalysisEvaluator {
private const string RandomParameterName = "Random";
private const string MultiVariateDataAnalysisProblemDataParameterName = "MultiVariateDataAnalysisProblemData";
private const string SymbolicExpressionTreeParameterName = "SymbolicExpressionTree";
private const string SymbolicExpressionTreeInterpreterParameterName = "SymbolicExpressionTreeInterpreter";
private const string SamplesStartParameterName = "SamplesStart";
private const string SamplesEndParameterName = "SamplesEnd";
private const string LowerEstimationLimitParameterName = "LowerEstimationLimit";
private const string UpperEstimationLimitParameterName = "UpperEstimationLimit";
private const string RelativeNumberOfEvaluatedSamplesParameterName = "RelativeNumberOfEvaluatedSamples";
#region parameter properties
public ILookupParameter RandomParameter {
get { return (ILookupParameter)Parameters[RandomParameterName]; }
}
public ILookupParameter MultiVariateDataAnalysisProblemDataParameter {
get { return (ILookupParameter)Parameters[MultiVariateDataAnalysisProblemDataParameterName]; }
}
public IValueLookupParameter SamplesStartParameter {
get { return (IValueLookupParameter)Parameters[SamplesStartParameterName]; }
}
public IValueLookupParameter SamplesEndParameter {
get { return (IValueLookupParameter)Parameters[SamplesEndParameterName]; }
}
public IValueLookupParameter LowerEstimationLimitParameter {
get { return (IValueLookupParameter)Parameters[LowerEstimationLimitParameterName]; }
}
public IValueLookupParameter UpperEstimationLimitParameter {
get { return (IValueLookupParameter)Parameters[UpperEstimationLimitParameterName]; }
}
public ILookupParameter SymbolicExpressionTreeParameter {
get { return (ILookupParameter)Parameters[SymbolicExpressionTreeParameterName]; }
}
public ILookupParameter SymbolicExpressionTreeInterpreterParameter {
get { return (ILookupParameter)Parameters[SymbolicExpressionTreeInterpreterParameterName]; }
}
public IValueParameter RelativeNumberOfEvaluatedSamplesParameter {
get { return (IValueParameter)Parameters[RelativeNumberOfEvaluatedSamplesParameterName]; }
}
#endregion
#region properties
public IRandom Random {
get { return RandomParameter.ActualValue; }
}
public ISymbolicExpressionTreeInterpreter SymbolicExpressionTreeInterpreter {
get { return SymbolicExpressionTreeInterpreterParameter.ActualValue; }
}
public SymbolicExpressionTree SymbolicExpressionTree {
get { return SymbolicExpressionTreeParameter.ActualValue; }
}
public MultiVariateDataAnalysisProblemData MultiVariateDataAnalysisProblemData {
get { return MultiVariateDataAnalysisProblemDataParameter.ActualValue; }
}
public IntValue SamplesStart {
get { return SamplesStartParameter.ActualValue; }
}
public IntValue SamplesEnd {
get { return SamplesEndParameter.ActualValue; }
}
public DoubleArray LowerEstimationLimit {
get { return LowerEstimationLimitParameter.ActualValue; }
}
public DoubleArray UpperEstimationLimit {
get { return UpperEstimationLimitParameter.ActualValue; }
}
public PercentValue RelativeNumberOfEvaluatedSamples {
get { return RelativeNumberOfEvaluatedSamplesParameter.Value; }
}
#endregion
public SymbolicVectorRegressionEvaluator()
: base() {
Parameters.Add(new LookupParameter(RandomParameterName, "A random number generator."));
Parameters.Add(new LookupParameter(MultiVariateDataAnalysisProblemDataParameterName, "The multi-variate data analysis problem data to use for training."));
Parameters.Add(new LookupParameter(SymbolicExpressionTreeInterpreterParameterName, "The tree interpreter that should be used to evaluate the symbolic expression tree."));
Parameters.Add(new ValueLookupParameter(SamplesStartParameterName, "The first index of the data set partition to use for training."));
Parameters.Add(new ValueLookupParameter(SamplesEndParameterName, "The last index of the data set partition to use for training."));
Parameters.Add(new ValueLookupParameter(UpperEstimationLimitParameterName, "The upper limit for the estimated values for each component."));
Parameters.Add(new ValueLookupParameter(LowerEstimationLimitParameterName, "The lower limit for the estimated values for each component."));
Parameters.Add(new LookupParameter(SymbolicExpressionTreeParameterName, "The symbolic vector regression solution encoded as a symbolic expression tree."));
Parameters.Add(new ValueParameter(RelativeNumberOfEvaluatedSamplesParameterName, "The relative number of samples of the dataset partition, which should be randomly chosen for evaluation between the start and end index.", new PercentValue(1)));
}
public override IOperation Apply() {
var interpreter = SymbolicExpressionTreeInterpreter;
var tree = SymbolicExpressionTree;
var problemData = MultiVariateDataAnalysisProblemData;
IEnumerable selectedTargetVariables =
problemData.TargetVariables.CheckedItems
.Select(x => x.Value.Value);
// check if there is a vector component for each target variable
if (selectedTargetVariables.Count() != tree.Root.SubTrees[0].SubTrees.Count)
throw new ArgumentException("The dimension of the output-vector of the tree doesn't match the number of selected target variables.");
int start = SamplesStart.Value;
int end = SamplesEnd.Value;
IEnumerable rows = GenerateRowsToEvaluate((uint)Random.Next(), RelativeNumberOfEvaluatedSamples.Value, start, end);
Evaluate(tree, interpreter, problemData, selectedTargetVariables, rows, LowerEstimationLimit, UpperEstimationLimit);
return base.Apply();
}
public abstract void Evaluate(SymbolicExpressionTree tree, ISymbolicExpressionTreeInterpreter interpreter, MultiVariateDataAnalysisProblemData problemData, IEnumerable targetVariables, IEnumerable rows, DoubleArray lowerEstimationBound, DoubleArray upperEstimationBound);
private static IEnumerable GenerateRowsToEvaluate(uint seed, double relativeAmount, int start, int end) {
if (end < start) throw new ArgumentException("Start value is larger than end value.");
int count = (int)((end - start) * relativeAmount);
if (count == 0) count = 1;
return RandomEnumerable.SampleRandomNumbers(seed, start, end, count);
}
}
}