[15744] | 1 | #region License Information
|
---|
| 2 | /* HeuristicLab
|
---|
| 3 | * Copyright (C) 2002-2018 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
|
---|
| 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 HeuristicLab.Common;
|
---|
| 26 | using HeuristicLab.Core;
|
---|
| 27 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
| 28 | using HeuristicLab.Problems.DataAnalysis;
|
---|
| 29 |
|
---|
| 30 | namespace HeuristicLab.Algorithms.DataAnalysis {
|
---|
| 31 | /// <summary>
|
---|
| 32 | /// Represents a linear regression model
|
---|
| 33 | /// </summary>
|
---|
| 34 | [StorableClass]
|
---|
| 35 | [Item("Linear Regression Model", "Represents a linear regression model.")]
|
---|
| 36 | public sealed class LinearRegressionModel : RegressionModel, IConfidenceRegressionModel {
|
---|
| 37 |
|
---|
| 38 | [Storable]
|
---|
| 39 | public double[,] C {
|
---|
| 40 | get; private set;
|
---|
| 41 | }
|
---|
| 42 | [Storable]
|
---|
| 43 | public double[] W {
|
---|
| 44 | get; private set;
|
---|
| 45 | }
|
---|
| 46 |
|
---|
| 47 | [Storable]
|
---|
| 48 | public double NoiseSigma {
|
---|
| 49 | get; private set;
|
---|
| 50 | }
|
---|
| 51 |
|
---|
| 52 | public override IEnumerable<string> VariablesUsedForPrediction {
|
---|
[16415] | 53 | get { return allowedInputVariables.Union(factorVariables.Select(f => f.Key)); }
|
---|
[15744] | 54 | }
|
---|
| 55 |
|
---|
| 56 | [Storable]
|
---|
| 57 | private string[] allowedInputVariables;
|
---|
| 58 | [Storable]
|
---|
| 59 | private List<KeyValuePair<string, IEnumerable<string>>> factorVariables;
|
---|
| 60 |
|
---|
| 61 | [StorableConstructor]
|
---|
| 62 | private LinearRegressionModel(bool deserializing)
|
---|
| 63 | : base(deserializing) {
|
---|
| 64 | }
|
---|
| 65 | private LinearRegressionModel(LinearRegressionModel original, Cloner cloner)
|
---|
| 66 | : base(original, cloner) {
|
---|
| 67 | this.W = original.W;
|
---|
| 68 | this.C = original.C;
|
---|
| 69 | this.NoiseSigma = original.NoiseSigma;
|
---|
| 70 |
|
---|
| 71 | allowedInputVariables = (string[])original.allowedInputVariables.Clone();
|
---|
| 72 | this.factorVariables = original.factorVariables.Select(kvp => new KeyValuePair<string, IEnumerable<string>>(kvp.Key, new List<string>(kvp.Value))).ToList();
|
---|
| 73 | }
|
---|
| 74 | public LinearRegressionModel(double[] w, double[,] covariance, double noiseSigma, string targetVariable, IEnumerable<string> doubleInputVariables, IEnumerable<KeyValuePair<string, IEnumerable<string>>> factorVariables)
|
---|
| 75 | : base(targetVariable) {
|
---|
| 76 | this.name = ItemName;
|
---|
| 77 | this.description = ItemDescription;
|
---|
| 78 | this.W = new double[w.Length];
|
---|
| 79 | Array.Copy(w, W, w.Length);
|
---|
| 80 | this.C = new double[covariance.GetLength(0),covariance.GetLength(1)];
|
---|
| 81 | Array.Copy(covariance, C, covariance.Length);
|
---|
| 82 | this.NoiseSigma = noiseSigma;
|
---|
[16415] | 83 | var stringInputVariables = factorVariables.Select(f => f.Key).Distinct();
|
---|
[15744] | 84 | this.allowedInputVariables = doubleInputVariables.ToArray();
|
---|
| 85 | this.factorVariables = factorVariables.Select(kvp => new KeyValuePair<string, IEnumerable<string>>(kvp.Key, new List<string>(kvp.Value))).ToList();
|
---|
| 86 | }
|
---|
| 87 |
|
---|
| 88 | [StorableHook(HookType.AfterDeserialization)]
|
---|
| 89 | private void AfterDeserialization() {
|
---|
| 90 | }
|
---|
| 91 |
|
---|
| 92 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
| 93 | return new LinearRegressionModel(this, cloner);
|
---|
| 94 | }
|
---|
| 95 |
|
---|
| 96 | public override IEnumerable<double> GetEstimatedValues(IDataset dataset, IEnumerable<int> rows) {
|
---|
| 97 | double[,] inputData = dataset.ToArray(allowedInputVariables, rows);
|
---|
| 98 | double[,] factorData = dataset.ToArray(factorVariables, rows);
|
---|
| 99 |
|
---|
| 100 | inputData = factorData.HorzCat(inputData);
|
---|
| 101 |
|
---|
| 102 | int n = inputData.GetLength(0);
|
---|
| 103 | int columns = inputData.GetLength(1);
|
---|
| 104 |
|
---|
| 105 | for (int row = 0; row < n; row++) {
|
---|
| 106 | double p = 0.0;
|
---|
| 107 | for (int column = 0; column < columns; column++) {
|
---|
| 108 | p += W[column] * inputData[row, column];
|
---|
| 109 | }
|
---|
| 110 | p += W[columns];
|
---|
| 111 | yield return p;
|
---|
| 112 | }
|
---|
| 113 | }
|
---|
| 114 |
|
---|
| 115 | public IEnumerable<double> GetEstimatedVariances(IDataset dataset, IEnumerable<int> rows) {
|
---|
| 116 | double[,] inputData = dataset.ToArray(allowedInputVariables, rows);
|
---|
| 117 | double[,] factorData = dataset.ToArray(factorVariables, rows);
|
---|
| 118 |
|
---|
| 119 | inputData = factorData.HorzCat(inputData);
|
---|
| 120 |
|
---|
| 121 | int n = inputData.GetLength(0);
|
---|
| 122 | int columns = inputData.GetLength(1);
|
---|
| 123 |
|
---|
| 124 | double[] d = new double[C.GetLength(0)];
|
---|
| 125 |
|
---|
| 126 | for (int row = 0; row < n; row++) {
|
---|
| 127 | for (int column = 0; column < columns; column++) {
|
---|
| 128 | d[column] = inputData[row,column];
|
---|
| 129 | }
|
---|
| 130 | d[columns] = 1;
|
---|
| 131 |
|
---|
| 132 | double var = 0.0;
|
---|
| 133 | for(int i=0;i<d.Length;i++) {
|
---|
| 134 | for(int j = 0;j<d.Length;j++) {
|
---|
| 135 | var += d[i] * C[i, j] * d[j];
|
---|
| 136 | }
|
---|
| 137 | }
|
---|
| 138 | yield return var + NoiseSigma*NoiseSigma;
|
---|
| 139 | }
|
---|
| 140 | }
|
---|
| 141 |
|
---|
| 142 |
|
---|
| 143 | public override IRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
|
---|
| 144 | return new ConfidenceRegressionSolution(this, new RegressionProblemData(problemData));
|
---|
| 145 | }
|
---|
| 146 | }
|
---|
| 147 | }
|
---|