svm各种工具箱 方法以后查找

NameAuthorInstitutionLanguageExecMulticlassRegressionComments

BSVMChih-Wei Hsu and Chih-Jen LinNational Taiwan UniversityC++WinYesYes 

Equbits ForesightEqubits LLCEqubits LLCSDKWin???YesCommercial. Contact Equibits LLC for details

Gini-SVMShantanu ChakrabarttyJohns Hopkins UniversityC++NoYesYesHandles non positive definite kernels

HeroSvmJianxiong DongConcordia UniversityC++WinYesNoOptimized for Pentium 4

LEARNSCVojislav Kecman Matlab p-filesN/AYesYesMust pay for source!

LIBSVMChih-Chung Chang, Chih-Jen LinNational Taiwan UniversityC++, Java, Python, R, MATLAB, Perl, RubyWin/*nixYesYesGraphic interface available

LS-SVMlabKristiaan Pelckmans, Johan SuykensKatholieke Universiteit LeuvenMatlabWin/*nixYesYesComes with platform-specific MEX files

Matlab SVM ToolboxS. R. GunnUniversity of SouthamptonMatlabN/ANoYesIncludes a simple GUI

mySVMStefan RupingUniversitat DortmundC++Win/*nixNoYes 

OSUJunshui Ma, Yi Zhao, and Stanley AhaltOhio State UniversityMatlabN/AYesYesMatlab interface to LIBSVM

Parallel GPDTT. Serafini, G. Zanghirati, L. ZanniUniversita di FerraraC++NoNoNoDesigned for parallel systems

pcSVM
Procoders.netC-NoNo 

RVMsMike TippingMSR CambridgeMatlab-YesYes 

SpiderSVMJason Weston, Andre Elisseeff , Gokhan BakIr , Fabian SinzMax Planck Institute for Biological CyberneticsMatlabN/AYesYesPart of the Spider machine learning library

Statistical Pattern Recognition Toolbox for MATLAB Vojtech Franc and Vaclav Hlavac Czech Technical University Prague Matlab/CNoYesNoGood online documentation. Everything I′ve tried has worked 字串9
well. Lots of stuff besides SVMs.

SVMdarkMartin SewellUniversity College LondonCWinNoYes 

SvmFuRyan RifkinMITC++No NoMust be compiled with g++

SVMLightThorsten JoachimsCornell UniversityCWin/*nixNoYes 

SVMsequelHal Daume IIIUniversity of Southern CaliforniaOCamlNoYesNo"Very fast and handles enormous datasets nicely"

SVMtorchRonan Collobert and Samy BengioIDIAPC++No?Yes 

SVM ToolboxGavin CawleyUniversity of East Anglia NorwichMatlab/C++NoYesNoBeta version

WinSVMMartin SewellUniversity College LondonC++WinNoYes