#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 .
*/
#endregion
using System;
using System.Collections.Generic;
using HeuristicLab.Core;
using HeuristicLab.Data;
using HeuristicLab.Operators;
using HeuristicLab.Random;
using System.Diagnostics;
namespace HeuristicLab.GP {
public class DeleteSubTreeManipulation : OperatorBase {
public override string Description {
get {
return @"Deletes a random sub-tree of the input tree. If the remaining tree is not valid
the operator tries to fix the tree by generating random subtrees where necessary.";
}
}
public DeleteSubTreeManipulation()
: base() {
AddVariableInfo(new VariableInfo("Random", "Uniform random number generator", typeof(MersenneTwister), VariableKind.In));
AddVariableInfo(new VariableInfo("OperatorLibrary", "The operator library containing all available operators", typeof(GPOperatorLibrary), VariableKind.In));
AddVariableInfo(new VariableInfo("MaxTreeHeight", "The maximal allowed height of the tree", typeof(IntData), VariableKind.In));
AddVariableInfo(new VariableInfo("MaxTreeSize", "The maximal allowed size (number of nodes) of the tree", typeof(IntData), VariableKind.In));
AddVariableInfo(new VariableInfo("FunctionTree", "The tree to mutate", typeof(IFunctionTree), VariableKind.In | VariableKind.Out));
AddVariableInfo(new VariableInfo("TreeSize", "The size (number of nodes) of the tree", typeof(IntData), VariableKind.In | VariableKind.Out));
AddVariableInfo(new VariableInfo("TreeHeight", "The height of the tree", typeof(IntData), VariableKind.In | VariableKind.Out));
}
public override IOperation Apply(IScope scope) {
IFunctionTree root = GetVariableValue("FunctionTree", scope, true);
MersenneTwister random = GetVariableValue("Random", scope, true);
GPOperatorLibrary library = GetVariableValue("OperatorLibrary", scope, true);
TreeGardener gardener = new TreeGardener(random, library);
IFunctionTree parent = gardener.GetRandomParentNode(root);
// parent==null means the whole tree should be deleted.
// => return a new minimal random tree
if(parent == null) {
IFunctionTree newTree = gardener.CreateBalancedRandomTree(1, 1);
// check if the tree is ok
Debug.Assert(gardener.IsValidTree(newTree));
// update sizes to match the new tree
GetVariableValue("TreeSize", scope, true).Data = newTree.Size;
GetVariableValue("TreeHeight", scope, true).Data = newTree.Height;
scope.GetVariable(scope.TranslateName("FunctionTree")).Value = newTree;
// schedule an operation to initialize the newly created operator
return gardener.CreateInitializationOperation(gardener.GetAllSubTrees(newTree), scope);
}
// select a branch to prune
int childIndex = random.Next(parent.SubTrees.Count);
if(parent.SubTrees.Count > parent.Function.MinArity) {
parent.RemoveSubTree(childIndex);
// actually since the next sub-trees are shifted in the place of the removed branch
// it might be possible that these sub-trees are not allowed in the place of the old branch
// we ignore this problem for now.
// when this starts to become a problem a possible solution is to go through the shifted branches from the place of the shifted
// and find the first one that doesn't fit. At this position we insert a new randomly initialized subtree of matching type (gkronber 25.12.07)
Debug.Assert(gardener.IsValidTree(root));
GetVariableValue("TreeSize", scope, true).Data = root.Size;
GetVariableValue("TreeHeight", scope, true).Data = root.Height;
// root hasn't changed so don't need to update 'FunctionTree' variable
return null;
} else {
// replace with a minimal random seedling
parent.RemoveSubTree(childIndex);
ICollection allowedFunctions = gardener.GetAllowedSubFunctions(parent.Function, childIndex);
IFunctionTree newFunctionTree = gardener.CreateRandomTree(allowedFunctions, 1, 1);
parent.InsertSubTree(childIndex, newFunctionTree);
Debug.Assert(gardener.IsValidTree(root));
GetVariableValue("TreeSize", scope, true).Data = root.Size;
GetVariableValue("TreeHeight", scope, true).Data = root.Height;
// again the root hasn't changed so we don't need to update the 'FunctionTree' variable
// but we have to return an initialization operation for the newly created tree
return gardener.CreateInitializationOperation(gardener.GetAllSubTrees(newFunctionTree), scope);
}
}
}
}