[2] | 1 | #region License Information
|
---|
| 2 | /* HeuristicLab
|
---|
| 3 | * Copyright (C) 2002-2008 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 System.Text;
|
---|
| 26 | using HeuristicLab.Core;
|
---|
| 27 | using HeuristicLab.Data;
|
---|
| 28 | using HeuristicLab.Operators;
|
---|
| 29 | using HeuristicLab.DataAnalysis;
|
---|
| 30 | using HeuristicLab.Functions;
|
---|
| 31 |
|
---|
| 32 | namespace HeuristicLab.StructureIdentification {
|
---|
[128] | 33 | public class VarianceAccountedForEvaluator : GPEvaluatorBase {
|
---|
[2] | 34 | public override string Description {
|
---|
[128] | 35 | get {
|
---|
[155] | 36 | return @"Evaluates 'FunctionTree' for all samples of 'DataSet' and calculates
|
---|
[2] | 37 | the variance-accounted-for quality measure for the estimated values vs. the real values of 'TargetVariable'.
|
---|
| 38 |
|
---|
| 39 | The Variance Accounted For (VAF) function is computed as
|
---|
| 40 | VAF(y,y') = ( 1 - var(y-y')/var(y) )
|
---|
[128] | 41 | where y' denotes the predicted / modelled values for y and var(x) the variance of a signal x.";
|
---|
| 42 | }
|
---|
[2] | 43 | }
|
---|
| 44 |
|
---|
| 45 | /// <summary>
|
---|
| 46 | /// The Variance Accounted For (VAF) function calculates is computed as
|
---|
| 47 | /// VAF(y,y') = ( 1 - var(y-y')/var(y) )
|
---|
| 48 | /// where y' denotes the predicted / modelled values for y and var(x) the variance of a signal x.
|
---|
| 49 | /// </summary>
|
---|
| 50 | public VarianceAccountedForEvaluator()
|
---|
| 51 | : base() {
|
---|
| 52 | }
|
---|
| 53 |
|
---|
| 54 |
|
---|
[155] | 55 | public override double Evaluate(IScope scope, IFunctionTree functionTree, int targetVariable, Dataset dataset) {
|
---|
[367] | 56 | int trainingStart = GetVariableValue<IntData>("TrainingSamplesStart", scope, true).Data;
|
---|
| 57 | int trainingEnd = GetVariableValue<IntData>("TrainingSamplesEnd", scope, true).Data;
|
---|
| 58 | double[] errors = new double[trainingEnd-trainingStart];
|
---|
| 59 | double[] originalTargetVariableValues = new double[trainingEnd-trainingStart];
|
---|
| 60 | double targetMean = dataset.GetMean(targetVariable, trainingStart, trainingEnd);
|
---|
[363] | 61 | functionTree.PrepareEvaluation(dataset);
|
---|
[367] | 62 | for(int sample = trainingStart; sample < trainingEnd; sample++) {
|
---|
[363] | 63 | double estimated = functionTree.Evaluate(sample);
|
---|
[128] | 64 | double original = dataset.GetValue(sample, targetVariable);
|
---|
[2] | 65 | if(!double.IsNaN(original) && !double.IsInfinity(original)) {
|
---|
| 66 | if(double.IsNaN(estimated) || double.IsInfinity(estimated))
|
---|
| 67 | estimated = targetMean + maximumPunishment;
|
---|
| 68 | else if(estimated > (targetMean + maximumPunishment))
|
---|
| 69 | estimated = targetMean + maximumPunishment;
|
---|
| 70 | else if(estimated < (targetMean - maximumPunishment))
|
---|
| 71 | estimated = targetMean - maximumPunishment;
|
---|
| 72 | }
|
---|
| 73 |
|
---|
[367] | 74 | errors[sample-trainingStart] = original - estimated;
|
---|
| 75 | originalTargetVariableValues[sample-trainingStart] = original;
|
---|
[2] | 76 | }
|
---|
| 77 |
|
---|
| 78 | double errorsVariance = Statistics.Variance(errors);
|
---|
| 79 | double originalsVariance = Statistics.Variance(originalTargetVariableValues);
|
---|
| 80 | double quality = 1 - errorsVariance / originalsVariance;
|
---|
| 81 |
|
---|
| 82 | if(double.IsNaN(quality) || double.IsInfinity(quality)) {
|
---|
| 83 | quality = double.MaxValue;
|
---|
| 84 | }
|
---|
[128] | 85 | return quality;
|
---|
[2] | 86 | }
|
---|
| 87 | }
|
---|
| 88 | }
|
---|