/// <summary>
/// Converts <see cref="SupportVectorMachineLearningConfigurationFunction"/>
/// into a lambda function that can be passed to the <see cref=" OneVsOneLearning{TInput, TBinary, TModel}.Learner"/>
/// property of a <see cref="MulticlassSupportVectorLearning"/> learning algorithm.
/// </summary>
///
public Func <InnerParameters, InnerLearning> Convert(SupportVectorMachineLearningConfigurationFunction conf)
{
return(delegate(InnerParameters parameters)
{
int[] y = Classes.ToMinusOnePlusOne(parameters.Outputs);
var machine = (KernelSupportVectorMachine)parameters.Model;
ISupportVectorMachineLearning r = conf(machine, parameters.Inputs, y, parameters.Pair.Class1, parameters.Pair.Class2);
var c = r as ISupervisedLearning <SupportVectorMachine <IKernel <double[]> >, double[], bool>;
if (c != null)
{
return c;
}
// TODO: The following checks exist only to provide support to previous way of using
// the library and should be removed after a few releases.
var svc = r as ISupportVectorMachineLearning;
if (svc != null)
{
svc.Run();
return null;
}
throw new Exception();
});
}
public void Reset()
{
// m_bow = null;
m_machine = null;
m_teacher = null;
m_trainImageFeatureVectors = null;
m_classIdClassNameMap = new Dictionary <int, string>();
}
/// <summary>
/// Learns a Support Vector Machines (SVM).
/// </summary>
/// <param name="trainer">The learning algorithm.</param>
/// <param name="x">The input vectors <paramref name="x"/>.</param>
/// <param name="y">The expected binary output <paramref name="y"/>.</param>
/// <param name="weights">The <c>weight</c> of importance for each input vector (if supported by the learning algorithm).</param>
/// <param name="cancellationToken">The cancellationToken token used to notify the machine that the operation should be canceled.</param>
/// <returns>
/// The <see cref="SupportVectorMachine"/> learned by this method.
/// A model that has learned how to produce <paramref name="y"/> given <paramref name="x"/>.
/// </returns>
public static SupportVectorMachine Learn(
ISupportVectorMachineLearning trainer,
IList <float[]> x,
IList <bool> y,
IList <float> weights,
CancellationToken cancellationToken)
{
if (trainer == null)
{
throw new ArgumentNullException(nameof(trainer));
}
return(trainer.Learn(x, y, weights, cancellationToken));
}
public static void train_one(Problem prob, Parameters param, out double[] w, double Cp, double Cn)
{
double[][] inputs = prob.Inputs;
int[] labels = prob.Outputs.Apply(x => x >= 0 ? 1 : -1);
double eps = param.Tolerance;
int pos = 0;
for (int i = 0; i < labels.Length; i++)
{
if (labels[i] >= 0)
{
pos++;
}
}
int neg = prob.Outputs.Length - pos;
double primal_solver_tol = eps * Math.Max(Math.Min(pos, neg), 1.0) / prob.Inputs.Length;
SupportVectorMachine svm = new SupportVectorMachine(prob.Dimensions);
ISupportVectorMachineLearning teacher = null;
switch (param.Solver)
{
case LibSvmSolverType.L2RegularizedLogisticRegression:
// l2r_lr_fun
teacher = new ProbabilisticNewtonMethod(svm, inputs, labels)
{
PositiveWeight = Cp,
NegativeWeight = Cn,
Tolerance = primal_solver_tol
}; break;
case LibSvmSolverType.L2RegularizedL2LossSvc:
// fun_obj=new l2r_l2_svc_fun(prob, C);
teacher = new LinearNewtonMethod(svm, inputs, labels)
{
PositiveWeight = Cp,
NegativeWeight = Cn,
Tolerance = primal_solver_tol
}; break;
case LibSvmSolverType.L2RegularizedL2LossSvcDual:
// solve_l2r_l1l2_svc(prob, w, eps, Cp, Cn, L2R_L2LOSS_SVC_DUAL);
teacher = new LinearDualCoordinateDescent(svm, inputs, labels)
{
Loss = Loss.L2,
PositiveWeight = Cp,
NegativeWeight = Cn,
}; break;
case LibSvmSolverType.L2RegularizedL1LossSvcDual:
// solve_l2r_l1l2_svc(prob, w, eps, Cp, Cn, L2R_L1LOSS_SVC_DUAL);
teacher = new LinearDualCoordinateDescent(svm, inputs, labels)
{
Loss = Loss.L1,
PositiveWeight = Cp,
NegativeWeight = Cn,
}; break;
case LibSvmSolverType.L1RegularizedLogisticRegression:
// solve_l1r_lr(&prob_col, w, primal_solver_tol, Cp, Cn);
teacher = new ProbabilisticCoordinateDescent(svm, inputs, labels)
{
PositiveWeight = Cp,
NegativeWeight = Cn,
Tolerance = primal_solver_tol
}; break;
case LibSvmSolverType.L2RegularizedLogisticRegressionDual:
// solve_l2r_lr_dual(prob, w, eps, Cp, Cn);
teacher = new ProbabilisticDualCoordinateDescent(svm, inputs, labels)
{
PositiveWeight = Cp,
NegativeWeight = Cn,
Tolerance = primal_solver_tol,
}; break;
}
Trace.WriteLine("Training " + param.Solver);
// run the learning algorithm
var sw = Stopwatch.StartNew();
double error = teacher.Run();
sw.Stop();
// save the solution
w = svm.ToWeights();
Trace.WriteLine(String.Format("Finished {0}: {1} in {2}",
param.Solver, error, sw.Elapsed));
}
/// <summary>
/// Learns a model that can map the given inputs to the given outputs.
/// </summary>
/// <param name="trainer">The learning algorithm.</param>
/// <param name="numberOfClasses">The number of classes.</param>
/// <param name="x">The input vectors <paramref name="x"/>.</param>
/// <param name="y">The expected binary output <paramref name="y"/>.</param>
/// <param name="weights">The <c>weight</c> of importance for each input vector (if supported by the learning algorithm).</param>
/// <param name="cancellationToken">The cancellationToken token used to notify the machine that the operation should be canceled.</param>
/// <exception cref="ArgumentNullException">
/// <para><paramref name="trainer"/> is <b>null</b>.</para>
/// <para>-or-</para>
/// <para><paramref name="x"/> is <b>null</b>.</para>
/// <para>-or-</para>
/// <para><paramref name="y"/> is <b>null</b>.</para>
/// </exception>
/// <exception cref="ArgumentException">
/// <para><paramref name="numberOfClasses"/> is less than 2.</para>
/// <para>-or-</para>
/// <para>The number of elements in <paramref name="y"/> does not match the number of elements in <paramref name="x"/>.</para>
/// <para>-or-</para>
/// <para><paramref name="weights"/> is not <b>null</b> and the number of elements in <paramref name="weights"/> does not match the number of elements in <paramref name="x"/>.</para>
/// </exception>
/// <returns>
/// The <see cref="OneVsAllSupportVectorMachine"/> learned by this method.
/// A model that has learned how to produce <paramref name="y"/> given <paramref name="x"/>.
/// </returns>
public static OneVsAllSupportVectorMachine Learn(
ISupportVectorMachineLearning trainer,
int numberOfClasses,
IList <float[]> x,
IList <int> y,
IList <float> weights,
CancellationToken cancellationToken)
{
if (trainer == null)
{
throw new ArgumentNullException(nameof(trainer));
}
if (x == null)
{
throw new ArgumentNullException(nameof(x));
}
if (y == null)
{
throw new ArgumentNullException(nameof(y));
}
if (numberOfClasses < 2)
{
throw new ArgumentException("The machine must have at least two classes.", nameof(numberOfClasses));
}
if (y.Count != x.Count)
{
throw new ArgumentException("The number of output labels must match the number of input vectors.", nameof(y));
}
// create the machines
SupportVectorMachine[] machines = new SupportVectorMachine[numberOfClasses];
// train each machine
int sampleCount = x.Count;
CommonParallel.For(
0,
machines.Length,
(a, b) =>
{
for (int i = a; i < b; i++)
{
bool[] expected = new bool[sampleCount];
for (int j = 0; j < sampleCount; j++)
{
expected[j] = y[j] == i;
}
machines[i] = SupportVectorMachine.Learn(trainer, x, expected, weights, cancellationToken);
}
},
new ParallelOptions()
{
CancellationToken = cancellationToken,
});
return(new OneVsAllSupportVectorMachine(machines));
}
// IKernel kernel;
public double BuildTheModel(double[][] inputs, int[] outputs, int ClassNum, ConfigurationFieldClassifier config)
{
cpuCounter.CategoryName = "Processor";
cpuCounter.CounterName = "% Processor Time";
cpuCounter.InstanceName = "_Total";
Reset();
_usenongoldenclass = config.FeatureExtraction.UseNonGoldenClass;
// scalers = scalresin;
IKernel kernal = null;
switch (config.AccordConfiguration.Kernel)
{
case KernelTypes.Gaussian:
kernal = new Gaussian(config.AccordConfiguration.GaussianKernel.Sigma);
break;
case KernelTypes.Polynomial:
kernal = new Polynomial(config.AccordConfiguration.PolynominalKernel.Degree, config.AccordConfiguration.PolynominalKernel.Constant);
break;
case KernelTypes.ChiSquare:
kernal = new ChiSquare();
break;
case KernelTypes.HistogramIntersction:
kernal = new HistogramIntersection();
break;
default:
break;
}
if (ClassNum > 2)
{
m_machine = new MulticlassSupportVectorMachine(inputs[0].Length, kernal, ClassNum);
m_teacher = (new MulticlassSupportVectorLearning((MulticlassSupportVectorMachine)m_machine, inputs, outputs));
(m_teacher as MulticlassSupportVectorLearning).Algorithm = (svm, classInputs, classOutputs, i, j) =>
{
var smo = new SequentialMinimalOptimization(svm, classInputs, classOutputs);
smo.Complexity = config.AccordConfiguration.Complexity;
smo.Tolerance = config.AccordConfiguration.Tolerance;
smo.CacheSize = config.AccordConfiguration.CacheSize;
smo.Strategy = (Accord.MachineLearning.VectorMachines.Learning.SelectionStrategy)((int)(config.AccordConfiguration.SelectionStrategy));
// smo.UseComplexityHeuristic = true;
// smo.PositiveWeight = 1;
int k = 0;
while (cpuCounter.NextValue() > 50)
{
Thread.Sleep(50);
k++;
if (k > 30000)
{
break;
}
}
// smo.NegativeWeight = 1;
smo.Run();
var probabilisticOutputLearning = new ProbabilisticOutputLearning(svm, classInputs, classOutputs);
return(probabilisticOutputLearning);
// return smo;
};
}
else
{
// FIX TO BASE TYPES THAN RUN THAN MAKE OTHER 2 CHANGES from LATEST - line ... and CLUSTER AND RUNTEST.. and check again...
m_machine = new SupportVectorMachine(inputs[0].Length);
m_teacher = new SequentialMinimalOptimization((SupportVectorMachine)m_machine, inputs, outputs);
(m_teacher as SequentialMinimalOptimization).Complexity = config.AccordConfiguration.Complexity;;
(m_teacher as SequentialMinimalOptimization).Tolerance = config.AccordConfiguration.Tolerance;
(m_teacher as SequentialMinimalOptimization).CacheSize = config.AccordConfiguration.CacheSize;
(m_teacher as SequentialMinimalOptimization).Strategy = (Accord.MachineLearning.VectorMachines.Learning.SelectionStrategy)((int)(config.AccordConfiguration.SelectionStrategy));
(m_teacher as SequentialMinimalOptimization).Complexity = config.AccordConfiguration.Complexity;;
}
// Configure the learning algorithm
// Train the machines. It should take a while.
// Thread.Sleep(10000);
//#if temp
double error = m_teacher.Run();
//#endif
// return 0;
return(error);
}
