source: branches/LearningClassifierSystems/HeuristicLab.Encodings.DecisionList/3.3/Rule.cs @ 10007

#region License Information
/* HeuristicLab
 * Copyright (C) 2002-2012 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 System.Text;
using HeuristicLab.Common;
using HeuristicLab.Core;
using HeuristicLab.Optimization.Operators.LCS;
using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;

namespace HeuristicLab.Encodings.DecisionList {
  [StorableClass]
  [Item("Rule", "")]
  public class Rule : Item {
    [Storable]
    private IDictionary<string, IVariable> variables;
    public IDictionary<string, IVariable> Variables {
      get { return variables; }
    }

    [Storable]
    private IAction action;
    public IAction Action { get { return action; } }

    public double Length {
      get { return variables.Values.Sum(x => x.Length); }
    }

    [StorableConstructor]
    protected Rule(bool deserializing) : base(deserializing) { }
    protected Rule(Rule original, Cloner cloner)
      : base(original, cloner) {
      variables = new Dictionary<string, IVariable>(original.variables.Count);
      foreach (var item in original.variables) {
        variables.Add(item.Key, cloner.Clone(item.Value));
      }
      action = (IAction)original.action.Clone();
    }
    public Rule()
      : base() {
      variables = new Dictionary<string, IVariable>();
    }
    public Rule(IEnumerable<IVariable> condition, IAction action)
      : this() {
      foreach (var variable in condition) {
        variables.Add(variable.VariableName, variable);
      }
      this.action = action;
    }
    public override IDeepCloneable Clone(Cloner cloner) {
      return new Rule(this, cloner);
    }

    public override string ToString() {
      StringBuilder sb = new StringBuilder();
      foreach (var variable in Variables) {
        sb.Append(variable.Value + "|");
      }
      sb.Append("|" + Action);
      return sb.ToString();
    }

    public void Randomize(IRandom random, double oneProbability, IEnumerable<IDiscretizer> discretizers, IEnumerable<IAction> exceptActions = null) {
      foreach (var variable in variables.Values) {
        variable.Randomize(random, oneProbability, discretizers);
      }
      if (exceptActions == null) {
        action.Randomize(random);
      } else {
        action.Randomize(random, exceptActions);
      }
    }

    public void SetToMatchInput(IGAssistInput input) {
      foreach (var variable in variables) {
        if (!input.VariableNames.Contains(variable.Key)) {
          throw new ArgumentException("input does not contain variable name of rule");
        }
        variable.Value.SetToMatch(input.GetVariableValue(variable.Key));
      }
    }

    public double ComputeTheoryLength() {
      return variables.Sum(x => x.Value.ComputeTheoryLength());
    }

    public bool MatchInput(IGAssistInput target) {
      foreach (var variable in variables) {
        if (!target.VariableNames.Contains(variable.Key)) { throw new ArgumentException("Input doesn't contain variable"); }
        if (!variable.Value.Match(target.GetVariableValue(variable.Key))) { return false; }
      }
      return true;
    }

    public Rule Crossover(Rule rule, IRandom random) {
      //variables and rule.Variables have to be the same
      int cutpoint = random.Next(0, Variables.Count);
      var crossedVariables = new List<IVariable>(variables.Count);
      var variableNames = variables.Keys.ToList();
      for (int i = 0; i < cutpoint; i++) {
        crossedVariables.Add(this.Variables[variableNames[i]]);
      }
      for (int i = cutpoint; i < variables.Count; i++) {
        crossedVariables.Add(rule.Variables[variableNames[i]]);
      }
      IAction action = random.Next(0, 2) == 0 ? this.action : rule.action;
      return new Rule(crossedVariables, action);
    }

    public void ApplySplit(IRandom random, double probability) {
      foreach (var variable in variables.Values) {
        if (random.NextDouble() < probability)
          variable.Split(random);
      }
    }

    public void ApplyMerge(IRandom random, double probability) {
      foreach (var variable in variables.Values) {
        if (random.NextDouble() < probability)
          variable.Merge(random);
      }
    }

    public void ApplyReinitialize(IRandom random, double probability, double oneProbability, IEnumerable<IDiscretizer> discretizers) {
      foreach (var variable in variables.Values) {
        if (random.NextDouble() < probability)
          variable.Reinitialize(random, oneProbability, discretizers);
      }
    }
  }
}