source: branches/2457_ExpertSystem/HeuristicLab.Problems.GeneralizedQuadraticAssignment/3.3/Operators/LocalImprovers/OneOptLocalSearch.cs @ 17175

#region License Information
/* HeuristicLab
 * Copyright (C) 2002-2018 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.Common;
using HeuristicLab.Core;
using HeuristicLab.Data;
using HeuristicLab.Encodings.IntegerVectorEncoding;
using HeuristicLab.Operators;
using HeuristicLab.Optimization;
using HeuristicLab.Parameters;
using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
using HeuristicLab.Random;
using HEAL.Attic;

namespace HeuristicLab.Problems.GeneralizedQuadraticAssignment {
  [Item("1-opt LocalSearch", "A simple exhaustive 1-change local search.")]
  [StorableType("C574770B-AB1E-46BE-8B3F-C8C5B2AD3ACF")]
  public class OneOptLocalSearch : SingleSuccessorOperator, IProblemInstanceAwareGQAPOperator,
    IQualityAwareGQAPOperator, IGQAPLocalImprovementOperator, IAssignmentAwareGQAPOperator, IStochasticOperator {
    public IProblem Problem { get; set; }
    public Type ProblemType {
      get { return typeof(GQAP); }
    }

    public ILookupParameter<GQAPInstance> ProblemInstanceParameter {
      get { return (ILookupParameter<GQAPInstance>)Parameters["ProblemInstance"]; }
    }
    public ILookupParameter<IntegerVector> AssignmentParameter {
      get { return (ILookupParameter<IntegerVector>)Parameters["Assignment"]; }
    }
    public ILookupParameter<DoubleValue> QualityParameter {
      get { return (ILookupParameter<DoubleValue>)Parameters["Quality"]; }
    }
    public ILookupParameter<Evaluation> EvaluationParameter {
      get { return (ILookupParameter<Evaluation>)Parameters["Evaluation"]; }
    }
    public IValueLookupParameter<IntValue> MaximumIterationsParameter {
      get { return (IValueLookupParameter<IntValue>)Parameters["MaximumIterations"]; }
    }
    public ILookupParameter<IntValue> EvaluatedSolutionsParameter {
      get { return (ILookupParameter<IntValue>)Parameters["EvaluatedSolutions"]; }
    }
    public ILookupParameter<IRandom> RandomParameter {
      get { return (ILookupParameter<IRandom>)Parameters["Random"]; }
    }
    public ILookupParameter<ResultCollection> ResultsParameter {
      get { return (ILookupParameter<ResultCollection>)Parameters["Results"]; }
    }

    [StorableConstructor]
    protected OneOptLocalSearch(StorableConstructorFlag _) : base(_) { }
    protected OneOptLocalSearch(OneOptLocalSearch original, Cloner cloner) : base(original, cloner) { }
    public OneOptLocalSearch()
      : base() {
      Parameters.Add(new LookupParameter<GQAPInstance>("ProblemInstance", GQAP.ProblemInstanceDescription));
      Parameters.Add(new LookupParameter<IntegerVector>("Assignment", GQAPSolutionCreator.AssignmentDescription));
      Parameters.Add(new LookupParameter<DoubleValue>("Quality", ""));
      Parameters.Add(new LookupParameter<Evaluation>("Evaluation", GQAP.EvaluationDescription));
      Parameters.Add(new ValueLookupParameter<IntValue>("MaximumIterations", "This parameter is ignored by the operator.") { Hidden = true });
      Parameters.Add(new LookupParameter<IntValue>("EvaluatedSolutions", "The number of evaluated solution equivalents."));
      Parameters.Add(new LookupParameter<IRandom>("Random", "The random number generator to use."));
      Parameters.Add(new LookupParameter<ResultCollection>("Results", "The result collection that stores the results."));
    }

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

    public static void Apply(IRandom random, GQAPSolution sol,
      GQAPInstance problemInstance, out int evaluatedSolutions) {
      var fit = problemInstance.ToSingleObjective(sol.Evaluation);
      var eval = sol.Evaluation;
      Apply(random, sol.Assignment, ref fit, ref eval, problemInstance, out evaluatedSolutions);
      sol.Evaluation = eval;
    }

    /// <summary>
    /// </summary>
    /// <param name="random">The random number generator to use.</param>
    /// <param name="assignment">The equipment-location assignment vector.</param>
    /// <param name="quality">The solution quality.</param>
    /// <param name="evaluation">The evaluation result of the solution.</param>
    /// <param name="maxCLS">The maximum number of candidates that should be found in each step.</param>
    /// <param name="oneMoveProbability">The probability for performing a 1-move, which is the opposite of performing a 2-move.</param>
    /// <param name="maximumIterations">The maximum number of iterations that should be performed each time the candidate list is generated.</param>
    /// <param name="problemInstance">The problem instance that contains the data.</param>
    /// <param name="evaluatedSolutions">The number of evaluated solutions.</param>
    /// <param name="greedy">Greedy selection performed better in 5 of 8 instances according to the paper</param>
    public static void Apply(IRandom random, IntegerVector assignment,
      ref double quality, ref Evaluation evaluation,
      GQAPInstance problemInstance, out int evaluatedSolutions) {
      evaluatedSolutions = 0;
      var capacities = problemInstance.Capacities;
      var demands = problemInstance.Demands;
      var evaluations = 0.0;
      var deltaEvaluationFactor = 1.0 / assignment.Length;
      var change = true;
      var moves = ExhaustiveOneMoveGenerator.GenerateAllNxM(problemInstance).ToList();
      var slack = evaluation.Slack.ToList();
      while (change) {
        change = false;
        var feasible = evaluation.IsFeasible;
        foreach (var move in moves
            .Where(x => {
              var equip = x.Indices[0];
              var assignTo = x.Reassignments[equip] - 1;
              if (assignTo == assignment[equip]) return false;
              var dem = problemInstance.Demands[equip];
              if (feasible) return slack[assignTo] >= dem;
              return slack[assignTo] - dem > slack[assignment[equip]];
            })
            .Shuffle(random)) {
          var moveEval = GQAPNMoveEvaluator.Evaluate(move, assignment, evaluation, problemInstance);
          evaluations += deltaEvaluationFactor;
          var moveQuality = problemInstance.ToSingleObjective(moveEval);
          if (moveQuality < quality) {
            NMoveMaker.Apply(assignment, move);
            quality = moveQuality;
            evaluation = moveEval;
            change = true;
            break;
          }
        }
      }
      evaluatedSolutions += (int)Math.Ceiling(evaluations);
    }

    private static NMove Move(IRandom random, IntegerVector assignment, double oneMoveProbability, DoubleArray capacities) {
      if (random.NextDouble() < oneMoveProbability)
        return StochasticNMoveSingleMoveGenerator.GenerateOneMove(random, assignment, capacities);
      return StochasticNMoveSingleMoveGenerator.GenerateTwoMove(random, assignment, capacities);
    }

    public override IOperation Apply() {
      var evaluation = EvaluationParameter.ActualValue;
      var quality = QualityParameter.ActualValue;
      var fit = quality.Value;
      var evaluatedSolutions = 0;

      Apply(RandomParameter.ActualValue,
        AssignmentParameter.ActualValue,
        ref fit,
        ref evaluation,
        ProblemInstanceParameter.ActualValue,
        out evaluatedSolutions);

      EvaluationParameter.ActualValue = evaluation;
      quality.Value = fit;
      EvaluatedSolutionsParameter.ActualValue.Value += evaluatedSolutions;
      return base.Apply();
    }
  }
}