[8401] | 1 | #region License Information
|
---|
| 2 | /* HeuristicLab
|
---|
| 3 | * Copyright (C) 2002-2012 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;
|
---|
[8366] | 23 | using HeuristicLab.Common;
|
---|
| 24 | using HeuristicLab.Core;
|
---|
| 25 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
| 26 |
|
---|
[8371] | 27 | namespace HeuristicLab.Algorithms.DataAnalysis {
|
---|
[8366] | 28 | [StorableClass]
|
---|
| 29 | [Item(Name = "CovarianceLinear", Description = "Linear covariance function with for Gaussian processes.")]
|
---|
| 30 | public class CovarianceLinear : Item, ICovarianceFunction {
|
---|
| 31 | [Storable]
|
---|
| 32 | private double[,] x;
|
---|
| 33 | [Storable]
|
---|
| 34 | private double[,] xt;
|
---|
| 35 |
|
---|
| 36 |
|
---|
| 37 | private double[,] k;
|
---|
| 38 | private bool symmetric;
|
---|
| 39 |
|
---|
| 40 | public int GetNumberOfParameters(int numberOfVariables) {
|
---|
| 41 | return 0;
|
---|
| 42 | }
|
---|
| 43 | [StorableConstructor]
|
---|
| 44 | protected CovarianceLinear(bool deserializing) : base(deserializing) { }
|
---|
| 45 | protected CovarianceLinear(CovarianceLinear original, Cloner cloner)
|
---|
| 46 | : base(original, cloner) {
|
---|
| 47 | // note: using shallow copies here!
|
---|
| 48 | this.x = original.x;
|
---|
| 49 | this.xt = original.xt;
|
---|
| 50 |
|
---|
| 51 | }
|
---|
| 52 | public CovarianceLinear()
|
---|
| 53 | : base() {
|
---|
| 54 | }
|
---|
| 55 |
|
---|
| 56 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
| 57 | return new CovarianceLinear(this, cloner);
|
---|
| 58 | }
|
---|
| 59 |
|
---|
| 60 | public void SetParameter(double[] hyp, double[,] x) {
|
---|
| 61 | if (hyp.Length > 0) throw new ArgumentException();
|
---|
| 62 | SetParameter(hyp, x, x);
|
---|
| 63 | this.symmetric = true;
|
---|
| 64 | }
|
---|
| 65 |
|
---|
| 66 | public void SetParameter(double[] hyp, double[,] x, double[,] xt) {
|
---|
| 67 | this.x = x;
|
---|
| 68 | this.xt = xt;
|
---|
| 69 | this.symmetric = false;
|
---|
| 70 |
|
---|
| 71 | k = null;
|
---|
| 72 | }
|
---|
| 73 |
|
---|
| 74 | public double GetCovariance(int i, int j) {
|
---|
| 75 | if (k == null) CalculateInnerProduct();
|
---|
| 76 | return k[i, j];
|
---|
| 77 | }
|
---|
| 78 |
|
---|
| 79 |
|
---|
| 80 | public double[] GetDiagonalCovariances() {
|
---|
| 81 | if (x != xt) throw new InvalidOperationException();
|
---|
| 82 | int rows = x.GetLength(0);
|
---|
| 83 | int cols = x.GetLength(1);
|
---|
| 84 | var k = new double[rows];
|
---|
| 85 | for (int i = 0; i < rows; i++) {
|
---|
| 86 | k[i] = 0;
|
---|
| 87 | for (int j = 0; j < cols; j++) {
|
---|
| 88 | k[i] += x[i, j] * x[i, j];
|
---|
| 89 | }
|
---|
| 90 | }
|
---|
| 91 | return k;
|
---|
| 92 | }
|
---|
| 93 |
|
---|
| 94 | public double[] GetGradient(int i, int j) {
|
---|
| 95 | throw new NotSupportedException();
|
---|
| 96 | }
|
---|
| 97 |
|
---|
| 98 |
|
---|
| 99 | private void CalculateInnerProduct() {
|
---|
| 100 | if (x.GetLength(1) != xt.GetLength(1)) throw new InvalidOperationException();
|
---|
| 101 | int rows = x.GetLength(0);
|
---|
| 102 | int cols = xt.GetLength(0);
|
---|
| 103 | k = new double[rows, cols];
|
---|
| 104 | if (symmetric) {
|
---|
| 105 | for (int i = 0; i < rows; i++) {
|
---|
| 106 | for (int j = i; j < cols; j++) {
|
---|
| 107 | k[i, j] = Util.ScalarProd(Util.GetRow(x, i),
|
---|
| 108 | Util.GetRow(x, j));
|
---|
| 109 | k[j, i] = k[i, j];
|
---|
| 110 | }
|
---|
| 111 | }
|
---|
| 112 | } else {
|
---|
| 113 | for (int i = 0; i < rows; i++) {
|
---|
| 114 | for (int j = 0; j < cols; j++) {
|
---|
| 115 | k[i, j] = Util.ScalarProd(Util.GetRow(x, i),
|
---|
| 116 | Util.GetRow(xt, j));
|
---|
| 117 | }
|
---|
| 118 | }
|
---|
| 119 | }
|
---|
| 120 | }
|
---|
| 121 | }
|
---|
| 122 | }
|
---|