LinearSVM: Linear SVM Classifier

In RSSL: Implementations of Semi-Supervised Learning Approaches for Classification

Linear SVM Classifier

Description

Implementation of the Linear Support Vector Classifier. Can be solved in the Dual formulation, which is equivalent to SVM or the Primal formulation.

Usage

LinearSVM(X, y, C = 1, method = "Dual", scale = TRUE, eps = 1e-09,
  reltol = 1e-13, maxit = 100)

Arguments

X	matrix; Design matrix for labeled data
y	factor or integer vector; Label vector
C	Cost variable
method	Estimation procedure c("Dual","Primal","BGD")
scale	Whether a z-transform should be applied (default: TRUE)
eps	Small value to ensure positive definiteness of the matrix in QP formulation
reltol	relative tolerance using during BFGS optimization
maxit	Maximum number of iterations for BFGS optimization

Value

S4 object of type LinearSVM

See Also

Other RSSL classifiers: EMLeastSquaresClassifier, EMLinearDiscriminantClassifier, GRFClassifier, ICLeastSquaresClassifier, ICLinearDiscriminantClassifier, KernelLeastSquaresClassifier, LaplacianKernelLeastSquaresClassifier(), LaplacianSVM, LeastSquaresClassifier, LinearDiscriminantClassifier, LinearTSVM(), LogisticLossClassifier, LogisticRegression, MCLinearDiscriminantClassifier, MCNearestMeanClassifier, MCPLDA, MajorityClassClassifier, NearestMeanClassifier, QuadraticDiscriminantClassifier, S4VM, SVM, SelfLearning, TSVM, USMLeastSquaresClassifier, WellSVM, svmlin()

RSSL documentation built on March 31, 2023, 7:27 p.m.