[7531] | 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 |
|
---|
[7562] | 22 | using System.Collections;
|
---|
[7531] | 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 |
|
---|
| 29 | namespace HeuristicLab.Problems.DataAnalysis {
|
---|
| 30 | [StorableClass]
|
---|
| 31 | [Item("MedianThresholdCalculator", "")]
|
---|
| 32 | public class MedianThresholdCalculator : DiscriminantClassificationWeightCalculator {
|
---|
| 33 |
|
---|
| 34 | public MedianThresholdCalculator()
|
---|
| 35 | : base() {
|
---|
| 36 | }
|
---|
| 37 | [StorableConstructor]
|
---|
| 38 | protected MedianThresholdCalculator(bool deserializing) : base(deserializing) { }
|
---|
| 39 | protected MedianThresholdCalculator(MedianThresholdCalculator original, Cloner cloner)
|
---|
| 40 | : base(original, cloner) {
|
---|
| 41 | }
|
---|
| 42 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
| 43 | return new MedianThresholdCalculator(this, cloner);
|
---|
| 44 | }
|
---|
| 45 |
|
---|
| 46 | protected double[] threshold;
|
---|
| 47 | protected double[] classValues;
|
---|
| 48 |
|
---|
[7549] | 49 | protected override IEnumerable<double> DiscriminantCalculateWeights(IEnumerable<IDiscriminantFunctionClassificationSolution> discriminantSolutions) {
|
---|
[7729] | 50 | classValues = discriminantSolutions.First().Model.ClassValues.ToArray();
|
---|
| 51 | var modelThresholds = discriminantSolutions.Select(x => x.Model.Thresholds.ToArray());
|
---|
| 52 | threshold = new double[modelThresholds.First().GetLength(0)];
|
---|
| 53 | for (int i = 0; i < modelThresholds.First().GetLength(0); i++) {
|
---|
| 54 | threshold[i] = GetMedian(modelThresholds.Select(x => x[i]).ToList());
|
---|
[7531] | 55 | }
|
---|
[7562] | 56 | return Enumerable.Repeat<double>(1, discriminantSolutions.Count());
|
---|
[7531] | 57 | }
|
---|
| 58 |
|
---|
[8814] | 59 | protected override double GetDiscriminantConfidence(IEnumerable<IDiscriminantFunctionClassificationSolution> solutions, int index, double estimatedClassValue, CheckPoint handler) {
|
---|
[7562] | 60 | Dataset dataset = solutions.First().ProblemData.Dataset;
|
---|
[8814] | 61 | IList<double> values = solutions.Where(s => handler(s.ProblemData, index)).Select(s => s.Model.GetEstimatedValues(dataset, Enumerable.Repeat(index, 1)).First()).ToList();
|
---|
[7562] | 62 | if (values.Count <= 0)
|
---|
| 63 | return double.NaN;
|
---|
| 64 | double median = GetMedian(values);
|
---|
[8297] | 65 | return GetMedianConfidence(median, estimatedClassValue);
|
---|
| 66 | }
|
---|
| 67 |
|
---|
[8814] | 68 | public override IEnumerable<double> GetDiscriminantConfidence(IEnumerable<IDiscriminantFunctionClassificationSolution> solutions, IEnumerable<int> indices, IEnumerable<double> estimatedClassValue, CheckPoint handler) {
|
---|
[8297] | 69 |
|
---|
| 70 | Dataset dataset = solutions.First().ProblemData.Dataset;
|
---|
[8814] | 71 | List<int> indicesList = indices.ToList();
|
---|
| 72 | var solValues = solutions.ToDictionary(x => x, x => x.Model.GetEstimatedValues(dataset, indicesList).ToArray());
|
---|
[8297] | 73 | double[] confidences = new double[indices.Count()];
|
---|
| 74 | double[] estimatedClassValueArr = estimatedClassValue.ToArray();
|
---|
| 75 |
|
---|
[8814] | 76 | for (int i = 0; i < indicesList.Count; i++) {
|
---|
| 77 | var values = solValues.Where(x => handler(x.Key.ProblemData, indicesList[i])).Select(x => x.Value[i]).ToList();
|
---|
| 78 | if (values.Count <= 0) {
|
---|
| 79 | confidences[i] = double.NaN;
|
---|
| 80 | } else {
|
---|
| 81 | double median = GetMedian(values);
|
---|
| 82 | confidences[i] = GetMedianConfidence(median, estimatedClassValueArr[i]);
|
---|
| 83 | }
|
---|
[8297] | 84 | }
|
---|
| 85 |
|
---|
| 86 | return confidences;
|
---|
| 87 | }
|
---|
| 88 |
|
---|
| 89 | protected double GetMedianConfidence(double median, double estimatedClassValue) {
|
---|
[8534] | 90 | for (int i = 0; i < classValues.Length; i++) {
|
---|
| 91 | if (estimatedClassValue.Equals(classValues[i])) {
|
---|
[8814] | 92 | //special case: median is higher than value of highest class
|
---|
| 93 | if (i == classValues.Length - 1 && median >= estimatedClassValue) {
|
---|
[8534] | 94 | return 1;
|
---|
| 95 | }
|
---|
[8814] | 96 | //special case: median is lower than value of lowest class
|
---|
[8534] | 97 | if (i == 0 && median < estimatedClassValue) {
|
---|
| 98 | return 1;
|
---|
| 99 | }
|
---|
[8814] | 100 | //special case: median is not between threshold of estimated class value
|
---|
[8534] | 101 | if ((i < classValues.Length - 1 && median >= threshold[i + 1]) || median <= threshold[i]) {
|
---|
| 102 | return 0;
|
---|
| 103 | }
|
---|
| 104 |
|
---|
[8814] | 105 | double thresholdToClassDistance, thresholdToMedianValueDistance;
|
---|
[8534] | 106 | if (median >= classValues[i]) {
|
---|
| 107 | thresholdToClassDistance = threshold[i + 1] - classValues[i];
|
---|
[8814] | 108 | thresholdToMedianValueDistance = threshold[i + 1] - median;
|
---|
[8534] | 109 | } else {
|
---|
| 110 | thresholdToClassDistance = classValues[i] - threshold[i];
|
---|
[8814] | 111 | thresholdToMedianValueDistance = median - threshold[i];
|
---|
[8534] | 112 | }
|
---|
[8814] | 113 | return (1 / thresholdToClassDistance) * thresholdToMedianValueDistance;
|
---|
[7562] | 114 | }
|
---|
[8534] | 115 | }
|
---|
| 116 | return double.NaN;
|
---|
[7562] | 117 | }
|
---|
| 118 |
|
---|
[8101] | 119 | protected double GetMedian(IList<double> estimatedValues) {
|
---|
[7531] | 120 | int count = estimatedValues.Count;
|
---|
| 121 | if (count % 2 == 0)
|
---|
[7549] | 122 | return 0.5 * (estimatedValues[count / 2 - 1] + estimatedValues[count / 2]);
|
---|
[7531] | 123 | else
|
---|
[7549] | 124 | return estimatedValues[count / 2];
|
---|
[7531] | 125 | }
|
---|
| 126 | }
|
---|
| 127 | }
|
---|