SPIDER The Spider Objects

SVM object

   SVM Support Vector Machine object             
   Generates a svm object with given hyperparameters.
     Hyperparameters (with defaults)
     child=kernel         -- the kernel is stored as a member called "child"
     C=Inf                -- the soft margin C parameter
     ridge=1e-13          -- a ridge on the kernel
     balanced_ridge=0     -- for unbalanced data
     nu = 0               -- Schoelkopf's nu svm parameter
     optimizer='default'  -- other choices={andre,quadprog,svmlight,
                                            libsvm,svmtorch(linux only)}
                             For "libsvm" optimizer you can specify the used cache size
                             by the global variable "libsvm_cachesize".
     alpha_cutoff=-1;     -- keep alphas with abs(a_i)>alpha_cutoff
                             default keeps all alphas, another
                             reasonable choice is e.g alpha_cutoff=1e-5 to remove
                             zero alphas (i.e non-SVs) to speed up computations.
      alpha               -- the weights
      b0                  -- the threshold
      Xsv                 -- the Support Vectors
    train, test, get_w 

Reference : A Tutorial on Support Vector Machines for Pattern Recognition
Author : Christopher J. C. Burges
Link : http://citeseer.ist.psu.edu/burges98tutorial.html
There is more than one svm available. See also help svm/svm.m