How to use libsvm based svm regression model for prediction in simulink ?

Dear Friends and Team, For one of my application, I am required to use libsvm based svm regression prediction model svmpredict in simulink. Currently I tried to achieve it through Simulink library MATLAB Function block, as shown in following code block

    function y = fcn(SOC,Curr,Tamb,switch_in)
    coder.extrinsic('svmpredict','net','polyvaln','evalin','assignin','mapminmax')
    ....
    switch switch_in
    case 1
        %'model' is the output structure of svmtrain and svmpredict is
        %external function
         assignin('base','xtestn',xtestn);
         estimate_out = svmpredict(y,xtestn, model); 
         disp(estimate_out);
        estimate_out = 0;
But using this the block goes to somekind of infinite simulation, without keeping track of the time necessary to complete simulation on time.

hence this is causing a stalled simulation at Matlab Function block.

Could anyone help me understand the problem or could anyone suggest me an alternative approach, to achieve the same.

Your help is grately appreciated.

Answer 1

[Ricardo Gualavisi]

Have you got any solution for this problem ? It would be very helpful for me. Thanks a lot!!