Which ls-svm toolbox can use in matlab? Which implementation do you recommend?
A:
Check out both libsvm and liblinear:
These the are the fastest SVM solvers that I know of.
carl
2010-06-14 04:32:15
I found libsvm can compute this type: 0 -- C-SVC 1 -- nu-SVC 2 -- one-class SVM 3 -- epsilon-SVR 4 -- nu-SVRCan you help, which one is least square?
Kukoda János
2010-06-14 10:01:56
A:
I am not 100% sure what type of SVM you're referring to, but I'm assuming you're interested in an implementation of the least squares SVM of Suykens & Vanderwalle, NIPS 99. If that's the case, I believe neither libsvm nor liblinear do that; check out http://www.esat.kuleuven.be/sista/lssvmlab/ .
If you're interested in a normal quadratic-programming formulation of the svm with quadratic slack penalties, libsvm and liblinear should work. Also, the newer subgradient-based solvers, such as pegasos may be useful as well but I am not sure of whether there is a good matlab library for you to use.
bsdfish
2010-06-17 22:47:57
I mixed up the soft margin SVM and LS-SVM. I looked for soft-margin SVM, not for LS-SVM. Since then i realized, that the standard MATLAB bioinfo svmtrain function can handle both. Thanks for everybody.
Kukoda János
2010-06-18 12:36:51