svmadmm: Linear/Nonlinear SVM Classification Solver Based on ADMM and IADMM Algorithms

Share:

Solve large-scale regularised linear/kernel classification by using ADMM and IADMM algorithms. This package provides linear L2-regularised primal classification (both ADMM and IADMM are available), kernel L2-regularised dual classification (IADMM) as well as L1-regularised primal classification (both ADMM and IADMM are available).

Author
Ben DAI <bendai2-c@my.cityu.edu.hk>; Junhui Wang <j.h.wang@cityu.edu.hk>
Date of publication
2016-03-29 10:50:29
Maintainer
Ben DAI <bendai2-c@my.cityu.edu.hk>
License
GPL-2
Version
0.3

View on CRAN

Man pages

svm.admm
linear/nonlinear svm solver based ADMM and IADMM algorithms
svm.predict
The predict function for svm.admm

Files in this package

svmadmm
svmadmm/src
svmadmm/src/kerneliadmm.c
svmadmm/src/linearadmm.c
svmadmm/src/lineariadmm.c
svmadmm/src/linear_math.c
svmadmm/src/linear_math.h
svmadmm/NAMESPACE
svmadmm/R
svmadmm/R/svm.predict.R
svmadmm/R/svm.admm.R
svmadmm/MD5
svmadmm/DESCRIPTION
svmadmm/man
svmadmm/man/svm.predict.Rd
svmadmm/man/svm.admm.Rd