[15830] | 1 | #region License Information
|
---|
| 2 | /* HeuristicLab
|
---|
[17209] | 3 | * Copyright (C) Heuristic and Evolutionary Algorithms Laboratory (HEAL)
|
---|
[15830] | 4 | *
|
---|
| 5 | * This file is part of HeuristicLab.
|
---|
| 6 | *
|
---|
| 7 | * HeuristicLab is free software: you can redistribute it and/or modify
|
---|
| 8 | * it under the terms of the GNU General Public License as published by
|
---|
| 9 | * the Free Software Foundation, either version 3 of the License, or
|
---|
| 10 | * (at your option) any later version.
|
---|
| 11 | *
|
---|
| 12 | * HeuristicLab is distributed in the hope that it will be useful,
|
---|
| 13 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 14 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 15 | * GNU General Public License for more details.
|
---|
| 16 | *
|
---|
| 17 | * You should have received a copy of the GNU General Public License
|
---|
| 18 | * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
|
---|
| 19 | */
|
---|
| 20 | #endregion
|
---|
| 21 |
|
---|
| 22 | using System;
|
---|
| 23 | using System.Collections.Generic;
|
---|
| 24 | using System.Linq;
|
---|
| 25 | using System.Threading;
|
---|
| 26 | using HeuristicLab.Common;
|
---|
| 27 | using HeuristicLab.Core;
|
---|
| 28 | using HeuristicLab.Optimization;
|
---|
| 29 | using HeuristicLab.Problems.DataAnalysis;
|
---|
[16847] | 30 | using HEAL.Attic;
|
---|
[15830] | 31 |
|
---|
| 32 | namespace HeuristicLab.Algorithms.DataAnalysis {
|
---|
[16847] | 33 | [StorableType("FAF1F955-82F3-4824-9759-9D2846E831AE")]
|
---|
[17081] | 34 | public class RegressionNodeTreeModel : RegressionModel, IDecisionTreeModel {
|
---|
[15830] | 35 | public const string NumCurrentLeafsResultName = "Number of current leafs";
|
---|
| 36 | public const string RootVariableName = "Root";
|
---|
| 37 | #region Properties
|
---|
| 38 | [Storable]
|
---|
| 39 | internal RegressionNodeModel Root { get; private set; }
|
---|
| 40 | #endregion
|
---|
| 41 |
|
---|
| 42 | #region HLConstructors & Cloning
|
---|
| 43 | [StorableConstructor]
|
---|
[16847] | 44 | protected RegressionNodeTreeModel(StorableConstructorFlag _) : base(_) { }
|
---|
[15830] | 45 | protected RegressionNodeTreeModel(RegressionNodeTreeModel original, Cloner cloner) : base(original, cloner) {
|
---|
| 46 | Root = cloner.Clone(original.Root);
|
---|
| 47 | }
|
---|
| 48 | protected RegressionNodeTreeModel(string targetVariable) : base(targetVariable) { }
|
---|
| 49 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
| 50 | return new RegressionNodeTreeModel(this, cloner);
|
---|
| 51 | }
|
---|
| 52 | #endregion
|
---|
| 53 |
|
---|
| 54 | internal static RegressionNodeTreeModel CreateTreeModel(string targetAttr, RegressionTreeParameters regressionTreeParams) {
|
---|
| 55 | return regressionTreeParams.LeafModel.ProvidesConfidence ? new ConfidenceRegressionNodeTreeModel(targetAttr) : new RegressionNodeTreeModel(targetAttr);
|
---|
| 56 | }
|
---|
| 57 |
|
---|
| 58 | #region RegressionModel
|
---|
| 59 | public override IEnumerable<string> VariablesUsedForPrediction {
|
---|
| 60 | get { return Root.VariablesUsedForPrediction ?? new List<string>(); }
|
---|
| 61 | }
|
---|
| 62 | public override IEnumerable<double> GetEstimatedValues(IDataset dataset, IEnumerable<int> rows) {
|
---|
| 63 | if (Root == null) throw new NotSupportedException("The model has not been built yet");
|
---|
| 64 | return Root.GetEstimatedValues(dataset, rows);
|
---|
| 65 | }
|
---|
| 66 | public override IRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
|
---|
| 67 | return new RegressionSolution(this, problemData);
|
---|
| 68 | }
|
---|
| 69 | #endregion
|
---|
| 70 |
|
---|
[17081] | 71 | #region IDecisionTreeModel
|
---|
[15830] | 72 | public void Build(IReadOnlyList<int> trainingRows, IReadOnlyList<int> pruningRows, IScope statescope, ResultCollection results, CancellationToken cancellationToken) {
|
---|
[17080] | 73 | var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
|
---|
[15830] | 74 | //start with one node
|
---|
| 75 | if (Root == null)
|
---|
| 76 | Root = RegressionNodeModel.CreateNode(regressionTreeParams.TargetVariable, regressionTreeParams);
|
---|
| 77 |
|
---|
| 78 | //split into (overfitted tree)
|
---|
| 79 | regressionTreeParams.Splitter.Split(this, trainingRows, statescope, cancellationToken);
|
---|
| 80 |
|
---|
| 81 | //prune
|
---|
| 82 | regressionTreeParams.Pruning.Prune(this, trainingRows, pruningRows, statescope, cancellationToken);
|
---|
| 83 |
|
---|
| 84 | //build final leaf models
|
---|
| 85 | regressionTreeParams.LeafModel.Build(this, trainingRows.Union(pruningRows).ToArray(), statescope, cancellationToken);
|
---|
| 86 | }
|
---|
| 87 |
|
---|
| 88 | public void Update(IReadOnlyList<int> rows, IScope statescope, CancellationToken cancellationToken) {
|
---|
[17080] | 89 | var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
|
---|
[15830] | 90 | regressionTreeParams.LeafModel.Build(this, rows, statescope, cancellationToken);
|
---|
| 91 | }
|
---|
| 92 |
|
---|
| 93 | public static void Initialize(IScope stateScope) {
|
---|
[17080] | 94 | var param = (RegressionTreeParameters)stateScope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
|
---|
[15830] | 95 | stateScope.Variables.Add(new Variable(RootVariableName, RegressionNodeModel.CreateNode(param.TargetVariable, param)));
|
---|
| 96 | }
|
---|
| 97 | #endregion
|
---|
| 98 |
|
---|
[16847] | 99 | public void BuildModel(IReadOnlyList<int> trainingRows, IReadOnlyList<int> pruningRows, IScope statescope, ResultCollection results, CancellationToken cancellationToken) {
|
---|
[17080] | 100 | var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
|
---|
[15830] | 101 | //start with one node
|
---|
| 102 | Root = RegressionNodeModel.CreateNode(regressionTreeParams.TargetVariable, regressionTreeParams);
|
---|
| 103 |
|
---|
| 104 | //split into (overfitted tree)
|
---|
| 105 | regressionTreeParams.Splitter.Split(this, trainingRows, statescope, cancellationToken);
|
---|
| 106 |
|
---|
| 107 | //prune
|
---|
| 108 | regressionTreeParams.Pruning.Prune(this, trainingRows, pruningRows, statescope, cancellationToken);
|
---|
| 109 | }
|
---|
| 110 |
|
---|
[16847] | 111 | [StorableType("E84ACC40-5694-4E40-A947-190673643206")]
|
---|
[15830] | 112 | private sealed class ConfidenceRegressionNodeTreeModel : RegressionNodeTreeModel, IConfidenceRegressionModel {
|
---|
| 113 | #region HLConstructors & Cloning
|
---|
| 114 | [StorableConstructor]
|
---|
[16847] | 115 | private ConfidenceRegressionNodeTreeModel(StorableConstructorFlag _) : base(_) { }
|
---|
[15830] | 116 | private ConfidenceRegressionNodeTreeModel(ConfidenceRegressionNodeTreeModel original, Cloner cloner) : base(original, cloner) { }
|
---|
| 117 | public ConfidenceRegressionNodeTreeModel(string targetVariable) : base(targetVariable) { }
|
---|
| 118 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
| 119 | return new ConfidenceRegressionNodeTreeModel(this, cloner);
|
---|
| 120 | }
|
---|
| 121 | #endregion
|
---|
| 122 |
|
---|
| 123 | public IEnumerable<double> GetEstimatedVariances(IDataset dataset, IEnumerable<int> rows) {
|
---|
| 124 | if (Root == null) throw new NotSupportedException("The model has not been built yet");
|
---|
| 125 | return ((IConfidenceRegressionModel)Root).GetEstimatedVariances(dataset, rows);
|
---|
| 126 | }
|
---|
| 127 | public override IRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
|
---|
| 128 | return new ConfidenceRegressionSolution(this, problemData);
|
---|
| 129 | }
|
---|
| 130 | }
|
---|
| 131 | }
|
---|
| 132 | } |
---|