source: trunk/sources/HeuristicLab.Random/UniformRandomAdder.cs @ 1094

#region License Information
/* HeuristicLab
 * Copyright (C) 2002-2008 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.Text;
using HeuristicLab.Core;
using HeuristicLab.Data;
using HeuristicLab.Constraints;

namespace HeuristicLab.Random {
  public class UniformRandomAdder : OperatorBase {

    private static int MAX_NUMBER_OF_TRIES = 100;

    public override string Description {
      get {
        return @"Samples a uniformly distributed random variable 'U' with range = [min,max] and E(u) = (max-min)/2
and adds the result to the variable 'Value'. ShakingFactor influences the effective range of U. 
If r=(max-min) then the effective range of U is [E(u) - shakingFactor * r/2, E(u) + shakingFactor * r/2].

If a constraint for the allowed range of 'Value' is defined and the result of the operation would be smaller then 
the smallest allowed value then 'Value' is set to the lower bound and vice versa for the upper bound.";
      }
    }

    public UniformRandomAdder() {
      AddVariableInfo(new VariableInfo("Value", "The value to manipulate (type is one of: IntData, ConstrainedIntData, DoubleData, ConstrainedDoubleData)", typeof(IObjectData), VariableKind.In));
      AddVariableInfo(new VariableInfo("ShakingFactor", "Determines the force of the shaking factor", typeof(DoubleData), VariableKind.In));
      AddVariableInfo(new VariableInfo("Random", "The random generator to use", typeof(MersenneTwister), VariableKind.In));
      AddVariableInfo(new VariableInfo("Min", "Lower bound of the uniform distribution (inclusive)", typeof(DoubleData), VariableKind.None));
      GetVariableInfo("Min").Local = true;
      AddVariable(new Variable("Min", new DoubleData(-1.0)));

      AddVariableInfo(new VariableInfo("Max", "Upper bound of the uniform distribution (exclusive)", typeof(DoubleData), VariableKind.None));
      GetVariableInfo("Max").Local = true;
      AddVariable(new Variable("Max", new DoubleData(1.0)));
    }

    public override IOperation Apply(IScope scope) {
      IObjectData value = GetVariableValue<IObjectData>("Value", scope, false);
      MersenneTwister mt = GetVariableValue<MersenneTwister>("Random", scope, true);
      double factor = GetVariableValue<DoubleData>("ShakingFactor", scope, true).Data;
      double min = GetVariableValue<DoubleData>("Min", scope, true).Data;
      double max = GetVariableValue<DoubleData>("Max", scope, true).Data;

      double ex = (max - min) / 2.0 + min;
      double newRange = (max - min) * factor;
      min = ex - newRange / 2;
      max = ex + newRange / 2;

      AddUniform(value, mt, min, max);
      return null;
    }

    private void AddUniform(IObjectData value, MersenneTwister mt, double min, double max) {
      // dispatch manually on dynamic type
      if (value is IntData)
        AddUniform((IntData)value, mt, min, max);
      else if (value is ConstrainedIntData)
        AddUniform((ConstrainedIntData)value, mt, min, max);
      else if (value is DoubleData)
        AddUniform((DoubleData)value, mt, min, max);
      else if (value is ConstrainedDoubleData)
        AddUniform((ConstrainedDoubleData)value, mt, min, max);
      else throw new InvalidOperationException("Can't handle type " + value.GetType().Name);
    }


    public void AddUniform(ConstrainedDoubleData data, MersenneTwister mt, double min, double max) {
      for (int tries = MAX_NUMBER_OF_TRIES; tries >= 0; tries--) {
        double newValue = data.Data + mt.NextDouble() * (max - min) + min;
        if (IsIntegerConstrained(data)) {
          newValue = Math.Floor(newValue);
        }
        if (data.TrySetData(newValue)) {
          return;
        }
      }
      throw new InvalidProgramException("Couldn't find a valid value");
    }

    public void AddUniform(ConstrainedIntData data, MersenneTwister mt, double min, double max) {
      for (int tries = MAX_NUMBER_OF_TRIES; tries >= 0; tries--) {
        int newValue = (int)Math.Floor(data.Data + mt.NextDouble() * (max - min) + min);
        if (data.TrySetData(newValue)) {
          return;
        }
      }
      throw new InvalidProgramException("Couldn't find a valid value");
    }

    public void AddUniform(DoubleData data, MersenneTwister mt, double min, double max) {
      data.Data = data.Data + mt.NextDouble() * (max - min) + min;
    }

    public void AddUniform(IntData data, MersenneTwister mt, double min, double max) {
      data.Data = (int)Math.Floor(data.Data + mt.NextDouble() * (max - min) + min);
    }
    private bool IsIntegerConstrained(ConstrainedDoubleData data) {
      foreach (IConstraint constraint in data.Constraints) {
        if (constraint is IsIntegerConstraint) {
          return true;
        }
      }
      return false;
    }
  }
}