#region License Information
/* HeuristicLab
* Copyright (C) 2002-2011 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
*
* This file is part of HeuristicLab.
*
* HeuristicLab is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* HeuristicLab is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with HeuristicLab. If not, see .
*/
#endregion
using System;
using System.Collections.Generic;
using System.Linq;
using HeuristicLab.Common;
using HeuristicLab.Core;
using HeuristicLab.Data;
using HeuristicLab.Optimization;
using HeuristicLab.Parameters;
using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
using HeuristicLab.Problems.DataAnalysis;
namespace HeuristicLab.Algorithms.DataAnalysis {
///
/// Support vector machine classification data analysis algorithm.
///
[Item("Support Vector Classification", "Support vector machine classification data analysis algorithm.")]
[Creatable("Data Analysis")]
[StorableClass]
public sealed class SupportVectorClassification : FixedDataAnalysisAlgorithm {
private const string SvmTypeParameterName = "SvmType";
private const string KernelTypeParameterName = "KernelType";
private const string CostParameterName = "Cost";
private const string NuParameterName = "Nu";
private const string GammaParameterName = "Gamma";
#region parameter properties
public IValueParameter SvmTypeParameter {
get { return (IValueParameter)Parameters[SvmTypeParameterName]; }
}
public IValueParameter KernelTypeParameter {
get { return (IValueParameter)Parameters[KernelTypeParameterName]; }
}
public IValueParameter NuParameter {
get { return (IValueParameter)Parameters[NuParameterName]; }
}
public IValueParameter CostParameter {
get { return (IValueParameter)Parameters[CostParameterName]; }
}
public IValueParameter GammaParameter {
get { return (IValueParameter)Parameters[GammaParameterName]; }
}
#endregion
#region properties
public StringValue SvmType {
get { return SvmTypeParameter.Value; }
}
public StringValue KernelType {
get { return KernelTypeParameter.Value; }
}
public DoubleValue Nu {
get { return NuParameter.Value; }
}
public DoubleValue Cost {
get { return CostParameter.Value; }
}
public DoubleValue Gamma {
get { return GammaParameter.Value; }
}
#endregion
[StorableConstructor]
private SupportVectorClassification(bool deserializing) : base(deserializing) { }
private SupportVectorClassification(SupportVectorClassification original, Cloner cloner)
: base(original, cloner) {
}
public SupportVectorClassification()
: base() {
Problem = new ClassificationProblem();
List svrTypes = (from type in new List { "NU_SVC", "EPSILON_SVC" }
select new StringValue(type).AsReadOnly())
.ToList();
ItemSet svrTypeSet = new ItemSet(svrTypes);
List kernelTypes = (from type in new List { "LINEAR", "POLY", "SIGMOID", "RBF" }
select new StringValue(type).AsReadOnly())
.ToList();
ItemSet kernelTypeSet = new ItemSet(kernelTypes);
Parameters.Add(new ConstrainedValueParameter(SvmTypeParameterName, "The type of SVM to use.", svrTypeSet, svrTypes[0]));
Parameters.Add(new ConstrainedValueParameter(KernelTypeParameterName, "The kernel type to use for the SVM.", kernelTypeSet, kernelTypes[3]));
Parameters.Add(new ValueParameter(NuParameterName, "The value of the nu parameter nu-SVC.", new DoubleValue(0.5)));
Parameters.Add(new ValueParameter(CostParameterName, "The value of the C (cost) parameter of C-SVC.", new DoubleValue(1.0)));
Parameters.Add(new ValueParameter(GammaParameterName, "The value of the gamma parameter in the kernel function.", new DoubleValue(1.0)));
}
[StorableHook(HookType.AfterDeserialization)]
private void AfterDeserialization() { }
public override IDeepCloneable Clone(Cloner cloner) {
return new SupportVectorClassification(this, cloner);
}
#region support vector classification
protected override void Run() {
IClassificationProblemData problemData = Problem.ProblemData;
IEnumerable selectedInputVariables = problemData.AllowedInputVariables;
var solution = CreateSupportVectorClassificationSolution(problemData, selectedInputVariables, SvmType.Value, KernelType.Value, Cost.Value, Nu.Value, Gamma.Value);
Results.Add(new Result("Support vector classification solution", "The support vector classification solution.", solution));
}
public static SupportVectorClassificationSolution CreateSupportVectorClassificationSolution(IClassificationProblemData problemData, IEnumerable allowedInputVariables,
string svmType, string kernelType, double cost, double nu, double gamma) {
Dataset dataset = problemData.Dataset;
string targetVariable = problemData.TargetVariable;
int start = problemData.TrainingPartition.Start;
int end = problemData.TrainingPartition.End;
IEnumerable rows = Enumerable.Range(start, end - start);
//extract SVM parameters from scope and set them
SVM.Parameter parameter = new SVM.Parameter();
parameter.SvmType = (SVM.SvmType)Enum.Parse(typeof(SVM.SvmType), svmType, true);
parameter.KernelType = (SVM.KernelType)Enum.Parse(typeof(SVM.KernelType), kernelType, true);
parameter.C = cost;
parameter.Nu = nu;
parameter.Gamma = gamma;
parameter.CacheSize = 500;
parameter.Probability = false;
SVM.Problem problem = SupportVectorMachineUtil.CreateSvmProblem(dataset, targetVariable, allowedInputVariables, rows);
SVM.RangeTransform rangeTransform = SVM.RangeTransform.Compute(problem);
SVM.Problem scaledProblem = SVM.Scaling.Scale(rangeTransform, problem);
var model = new SupportVectorMachineModel(SVM.Training.Train(scaledProblem, parameter), rangeTransform, targetVariable, allowedInputVariables, problemData.ClassValues);
return new SupportVectorClassificationSolution(model, (IClassificationProblemData)problemData.Clone());
}
#endregion
}
}