LinearTSVM {RSSL}

Linear CCCP Transductive SVM classifier

Description

Implementation for the Linear TSVM. This method is mostly for debugging purposes and does not allow for the balancing constraint or kernels, like the TSVM function.

Usage

LinearTSVM(X, y, X_u, C, Cstar, s = 0, x_center = FALSE, scale = FALSE,
  eps = 1e-06, verbose = FALSE, init = NULL)

Arguments

X

matrix; Design matrix, intercept term is added within the function

y

vector; Vector or factor with class assignments

X_u

matrix; Design matrix of the unlabeled data, intercept term is added within the function

C

numeric; Cost parameter of the SVM

Cstar

numeric; Cost parameter of the unlabeled objects

s

numeric; parameter controlling the loss function of the unlabeled objects

x_center

logical; Should the features be centered?

scale

logical; If TRUE, apply a z-transform to all observations in X and X_u before running the regression

eps

numeric; Convergence criterion

verbose

logical; print debugging messages (default: FALSE)

init

numeric; Initial classifier parameters to start the convex concave procedure

References

Collobert, R. et al., 2006. Large scale transductive SVMs. Journal of Machine Learning Research, 7, pp.1687-1712.

See Also

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