[6567] | 1 | #region License Information |
---|
| 2 | /* HeuristicLab |
---|
[14185] | 3 | * Copyright (C) 2002-2016 Heuristic and Evolutionary Algorithms Laboratory (HEAL) |
---|
[6567] | 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; |
---|
[14542] | 25 | using System.Threading; |
---|
[6567] | 26 | using HeuristicLab.Common; |
---|
| 27 | using HeuristicLab.Core; |
---|
| 28 | using HeuristicLab.Data; |
---|
| 29 | using HeuristicLab.Optimization; |
---|
| 30 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable; |
---|
| 31 | using HeuristicLab.Problems.DataAnalysis; |
---|
| 32 | |
---|
| 33 | namespace HeuristicLab.Algorithms.DataAnalysis { |
---|
| 34 | /// <summary> |
---|
| 35 | /// Multinomial logit regression data analysis algorithm. |
---|
| 36 | /// </summary> |
---|
[13238] | 37 | [Item("Multinomial Logit Classification (MNL)", "Multinomial logit classification data analysis algorithm (wrapper for ALGLIB).")] |
---|
[12622] | 38 | [Creatable(CreatableAttribute.Categories.DataAnalysisClassification, Priority = 180)] |
---|
[6567] | 39 | [StorableClass] |
---|
| 40 | public sealed class MultiNomialLogitClassification : FixedDataAnalysisAlgorithm<IClassificationProblem> { |
---|
| 41 | private const string LogitClassificationModelResultName = "Logit classification solution"; |
---|
| 42 | |
---|
| 43 | [StorableConstructor] |
---|
| 44 | private MultiNomialLogitClassification(bool deserializing) : base(deserializing) { } |
---|
| 45 | private MultiNomialLogitClassification(MultiNomialLogitClassification original, Cloner cloner) |
---|
| 46 | : base(original, cloner) { |
---|
| 47 | } |
---|
| 48 | public MultiNomialLogitClassification() |
---|
| 49 | : base() { |
---|
| 50 | Problem = new ClassificationProblem(); |
---|
| 51 | } |
---|
| 52 | [StorableHook(HookType.AfterDeserialization)] |
---|
| 53 | private void AfterDeserialization() { } |
---|
| 54 | |
---|
| 55 | public override IDeepCloneable Clone(Cloner cloner) { |
---|
| 56 | return new MultiNomialLogitClassification(this, cloner); |
---|
| 57 | } |
---|
| 58 | |
---|
[6633] | 59 | #region logit classification |
---|
[14542] | 60 | protected override void Run(CancellationToken cancellationToken) { |
---|
[6567] | 61 | double rmsError, relClassError; |
---|
| 62 | var solution = CreateLogitClassificationSolution(Problem.ProblemData, out rmsError, out relClassError); |
---|
[6633] | 63 | Results.Add(new Result(LogitClassificationModelResultName, "The logit classification solution.", solution)); |
---|
| 64 | Results.Add(new Result("Root mean squared error", "The root of the mean of squared errors of the logit regression solution on the training set.", new DoubleValue(rmsError))); |
---|
[6567] | 65 | Results.Add(new Result("Relative classification error", "Relative classification error on the training set (percentage of misclassified cases).", new PercentValue(relClassError))); |
---|
| 66 | } |
---|
| 67 | |
---|
| 68 | public static IClassificationSolution CreateLogitClassificationSolution(IClassificationProblemData problemData, out double rmsError, out double relClassError) { |
---|
[12509] | 69 | var dataset = problemData.Dataset; |
---|
[6567] | 70 | string targetVariable = problemData.TargetVariable; |
---|
[14240] | 71 | var doubleVariableNames = problemData.AllowedInputVariables.Where(dataset.VariableHasType<double>); |
---|
| 72 | var factorVariableNames = problemData.AllowedInputVariables.Where(dataset.VariableHasType<string>); |
---|
[8139] | 73 | IEnumerable<int> rows = problemData.TrainingIndices; |
---|
[14240] | 74 | double[,] inputMatrix = AlglibUtil.PrepareInputMatrix(dataset, doubleVariableNames.Concat(new string[] { targetVariable }), rows); |
---|
| 75 | |
---|
| 76 | var factorVariableValues = AlglibUtil.GetFactorVariableValues(dataset, factorVariableNames, rows); |
---|
| 77 | var factorMatrix = AlglibUtil.PrepareInputMatrix(dataset, factorVariableValues, rows); |
---|
| 78 | inputMatrix = factorMatrix.VertCat(inputMatrix); |
---|
| 79 | |
---|
[6567] | 80 | if (inputMatrix.Cast<double>().Any(x => double.IsNaN(x) || double.IsInfinity(x))) |
---|
| 81 | throw new NotSupportedException("Multinomial logit classification does not support NaN or infinity values in the input dataset."); |
---|
| 82 | |
---|
[12817] | 83 | alglib.logitmodel lm = new alglib.logitmodel(); |
---|
| 84 | alglib.mnlreport rep = new alglib.mnlreport(); |
---|
[6567] | 85 | int nRows = inputMatrix.GetLength(0); |
---|
| 86 | int nFeatures = inputMatrix.GetLength(1) - 1; |
---|
[6740] | 87 | double[] classValues = dataset.GetDoubleValues(targetVariable).Distinct().OrderBy(x => x).ToArray(); |
---|
[6567] | 88 | int nClasses = classValues.Count(); |
---|
| 89 | // map original class values to values [0..nClasses-1] |
---|
[8139] | 90 | Dictionary<double, double> classIndices = new Dictionary<double, double>(); |
---|
[6567] | 91 | for (int i = 0; i < nClasses; i++) { |
---|
[8139] | 92 | classIndices[classValues[i]] = i; |
---|
[6567] | 93 | } |
---|
| 94 | for (int row = 0; row < nRows; row++) { |
---|
[8139] | 95 | inputMatrix[row, nFeatures] = classIndices[inputMatrix[row, nFeatures]]; |
---|
[6567] | 96 | } |
---|
| 97 | int info; |
---|
| 98 | alglib.mnltrainh(inputMatrix, nRows, nFeatures, nClasses, out info, out lm, out rep); |
---|
| 99 | if (info != 1) throw new ArgumentException("Error in calculation of logit classification solution"); |
---|
| 100 | |
---|
| 101 | rmsError = alglib.mnlrmserror(lm, inputMatrix, nRows); |
---|
| 102 | relClassError = alglib.mnlrelclserror(lm, inputMatrix, nRows); |
---|
| 103 | |
---|
[14240] | 104 | MultinomialLogitClassificationSolution solution = new MultinomialLogitClassificationSolution(new MultinomialLogitModel(lm, targetVariable, doubleVariableNames, factorVariableValues, classValues), (IClassificationProblemData)problemData.Clone()); |
---|
[6567] | 105 | return solution; |
---|
| 106 | } |
---|
| 107 | #endregion |
---|
| 108 | } |
---|
| 109 | } |
---|