public override IModelLikelihood <double, int> GenerateModelLikelihood(IDataSet <double, int> training_set)
{
svm_problem prob = new svm_problem();
prob.l = training_set.InstanceCount;
prob.x = CreateNodeArray(ToolsCollection.ConvertToArray2D(training_set.FeatureData));
prob.y = ToolsCollection.ConvertToDoubleArray(ToolsCollection.ConvertToArray2D(training_set.LabelData).Select1DIndex1(0));
//Train model---------------------------------------------------------------------
return(new ModelLibSVMCSVC(training_set.DataContext, new C_SVC(prob, KernelHelper.RadialBasisFunctionKernel(this.Gamma), this.C, this.CacheSize, true)));
}
public AMatrix <Matrix <double> > Create(float[,] operant_0)
{
return(Create(ToolsCollection.ConvertToDoubleArray(operant_0)));
}
private DenseMatrix CreateDenseMatrix(AMatrix <Matrix <double> > operant_0, IList <double> operant_1)
{
return(new DenseMatrix(operant_0.Data.RowCount, operant_0.Data.ColumnCount, ToolsCollection.ConvertToDoubleArray(operant_1)));
}
public AMatrix <Matrix <double> > Create(IList <float> operant_0, bool transpose = false)
{
return(Create(ToolsCollection.ConvertToDoubleArray(operant_0), transpose));
}
public void AddSeries(IList <float> x_values, IList <float> y_values)
{
AddSeries(ToolsCollection.ConvertToDoubleArray(x_values), ToolsCollection.ConvertToDoubleArray(y_values), Color.Black);
}
public float [] ComputeInverse(float [] input)
{
return(((matrix_backward * ToolsCollection.ConvertToDoubleArray(input)) + means).ToArray1DFloat32());
}
public float [] Compute(float [] input)
{
return((matrix_forward * (ToolsCollection.ConvertToDoubleArray(input) - means)).ToArray1DFloat32());
}
