MATLAB fitcSVM weight vector

Question

I am training a linear SVM classifier with the fitcsvm function in MATLAB:

     cvFolds = crossvalind('Kfold', labels, nrFolds);

          for i = 1:nrFolds                       % iterate through each fold

              testIdx = (cvFolds == i);            % indices of test instances
              trainIdx = ~testIdx;                 % indices training instances

              cl = fitcsvm(features(trainIdx,:), 
              labels(trainIdx),'KernelFunction',kernel,'Standardize',true,...
             'BoxConstraint',C,'ClassNames',[0,1], 'Solver', solver);

              [labelPred,scores] =  predict(cl, features(testIdx,:));
              eq = sum(labelPred==labels(testIdx));
              accuracy(i) = eq/numel(labels(testIdx));

          end

As visible from this part of code, the trained SVM model is stored in cl. Checking the model parameters in cl I do not see which parameters correspond to classifier weight - ie. the parameter for linear classifiers which reflects the importance of each feature. Which parameter represents the classification weights? I see in the MATLAB documentation "The vector β contains the coefficients that define an orthogonal vector to the hyperplane" - is hence cl.beta representing the classification weights?

Answer 1

As you can see in this documentation, the equation of a hyperplane in fitcsvm is

                                 f(x)=x′β+b=0

And as you know, this equation shows following relationship:

                                 f(x)=w*x+b=0 or f(x)=x*w+b=0

So, β is equal to w (weights).