wellsvm_direct {RSSL}

wellsvm implements the wellsvm algorithm as shown in [1].

Description

wellsvm implements the wellsvm algorithm as shown in [1].

Usage

wellsvm_direct(x, y, testx, testy, C1 = 1, C2 = 0.1, gamma = 1)

Arguments

x

A Nxd training data matrix, where N is the number of training instances and d is the dimension of instance;

y

A Nx1 training label vector, where y = 1/-1 means positive/negative, and y = 0 means unlabeled;

testx

A Mxd testing data matrix, where M is the number of testing instances;

testy

A Mx1 testing label vector

C1

A regularization parameter for labeled data, default 1;

C2

A regularization parameter for unlabeled data, default 0.1;

gamma

Gaussian kernel parameter, i.e., k(x,y) = exp(-gamma^2||x-y||^2/avg) where avg is the average distance among instances; when gamma = 0, linear kernel is used. default gamma = 1;

Value

prediction - A Mx1 predicted testing label vector; accuracy - The accuracy of prediction; cputime - cpu running time;

References

Y.-F. Li, I. W. Tsang, J. T. Kwok, and Z.-H. Zhou. Scalable and Convex Weakly Labeled SVMs. Journal of Machine Learning Research, 2013.

R.-E. Fan, P.-H. Chen, and C.-J. Lin. Working set selection using second order information for training SVM. Journal of Machine Learning Research 6, 1889-1918, 2005.