[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 HeuristicLab.Common;
|
---|
| 26 |
|
---|
| 27 | namespace HeuristicLab.Algorithms.DataAnalysis {
|
---|
| 28 | /// <summary>
|
---|
| 29 | /// Helper class for incremental split calculation.
|
---|
[16847] | 30 | /// Used while moving a potential splitter along the ordered training instances
|
---|
[15830] | 31 | /// </summary>
|
---|
| 32 | internal class UnivariateOnlineLR {
|
---|
| 33 | #region state
|
---|
| 34 | private readonly NeumaierSum targetMean;
|
---|
| 35 | private readonly NeumaierSum attributeMean;
|
---|
| 36 | private readonly NeumaierSum targetVarSum;
|
---|
| 37 | private readonly NeumaierSum attributeVarSum;
|
---|
| 38 | private readonly NeumaierSum comoment;
|
---|
| 39 | private readonly NeumaierSum ssr;
|
---|
| 40 | private int size;
|
---|
| 41 | #endregion
|
---|
| 42 |
|
---|
| 43 | public double Ssr {
|
---|
| 44 | get { return ssr.Get(); }
|
---|
| 45 | }
|
---|
| 46 | public int Size {
|
---|
| 47 | get { return size; }
|
---|
| 48 | }
|
---|
| 49 |
|
---|
| 50 | private double Beta {
|
---|
| 51 | get { return comoment.Get() / attributeVarSum.Get(); }
|
---|
| 52 | }
|
---|
| 53 | private double Alpha {
|
---|
| 54 | get { return targetMean.Get() - Beta * attributeMean.Get(); }
|
---|
| 55 | }
|
---|
| 56 |
|
---|
| 57 | public UnivariateOnlineLR(ICollection<double> attributeValues, ICollection<double> targetValues) {
|
---|
| 58 | if (attributeValues.Count != targetValues.Count) throw new ArgumentException("Targets and Attributes need to have the same length");
|
---|
| 59 | size = attributeValues.Count;
|
---|
| 60 |
|
---|
| 61 | var yMean = targetValues.Average();
|
---|
| 62 | var xMean = attributeValues.Average();
|
---|
| 63 | targetMean = new NeumaierSum(yMean);
|
---|
| 64 | attributeMean = new NeumaierSum(xMean);
|
---|
| 65 | targetVarSum = new NeumaierSum(targetValues.VariancePop() * size);
|
---|
| 66 | attributeVarSum = new NeumaierSum(attributeValues.VariancePop() * size);
|
---|
| 67 | comoment = new NeumaierSum(attributeValues.Zip(targetValues, (x, y) => (x - xMean) * (y - yMean)).Sum());
|
---|
| 68 |
|
---|
| 69 | var beta = comoment.Get() / attributeVarSum.Get();
|
---|
| 70 | var alpha = yMean - beta * xMean;
|
---|
| 71 | ssr = new NeumaierSum(attributeValues.Zip(targetValues, (x, y) => y - alpha - beta * x).Sum(x => x * x));
|
---|
| 72 | }
|
---|
| 73 |
|
---|
| 74 | public void Add(double attributeValue, double targetValue) {
|
---|
| 75 | var predictOld = Predict(attributeValue, targetValue);
|
---|
| 76 |
|
---|
| 77 | size++;
|
---|
| 78 | var dx = attributeValue - attributeMean.Get();
|
---|
| 79 | var dy = targetValue - targetMean.Get();
|
---|
| 80 | attributeMean.Add(dx / size);
|
---|
| 81 | targetMean.Add(dy / size);
|
---|
| 82 | var dx2 = attributeValue - attributeMean.Get();
|
---|
| 83 | var dy2 = targetValue - targetMean.Get();
|
---|
| 84 | attributeVarSum.Add(dx * dx2);
|
---|
| 85 | targetVarSum.Add(dy * dy2);
|
---|
| 86 | comoment.Add(dx * dy2);
|
---|
| 87 |
|
---|
| 88 | ssr.Add(predictOld * Predict(attributeValue, targetValue));
|
---|
| 89 | }
|
---|
| 90 |
|
---|
| 91 | public void Remove(double attributeValue, double targetValue) {
|
---|
| 92 | var predictOld = Predict(attributeValue, targetValue);
|
---|
| 93 |
|
---|
| 94 | var dx2 = attributeValue - attributeMean.Get();
|
---|
| 95 | var dy2 = targetValue - targetMean.Get();
|
---|
| 96 | attributeMean.Mul(size / (size - 1.0));
|
---|
| 97 | targetMean.Mul(size / (size - 1.0));
|
---|
| 98 | attributeMean.Add(-attributeValue / (size - 1.0));
|
---|
| 99 | targetMean.Add(-targetValue / (size - 1.0));
|
---|
| 100 | var dx = attributeValue - attributeMean.Get();
|
---|
| 101 | var dy = targetValue - targetMean.Get();
|
---|
| 102 | attributeVarSum.Add(-dx * dx2);
|
---|
| 103 | targetVarSum.Add(-dy * dy2);
|
---|
| 104 | comoment.Add(-dx * dy2);
|
---|
| 105 | size--;
|
---|
| 106 |
|
---|
| 107 | ssr.Add(-predictOld * Predict(attributeValue, targetValue));
|
---|
| 108 | }
|
---|
| 109 |
|
---|
| 110 | private double Predict(double attributeValue, double targetValue) {
|
---|
| 111 | return targetValue - Alpha - Beta * attributeValue;
|
---|
| 112 | }
|
---|
| 113 | }
|
---|
| 114 | } |
---|