SVM creation using struct variable

Hi, I am struggling trying to create a good SVM. Sadly, while my sensitivity is around 92 specificity is low, about 48 I'd like to balance them, so I am trying to manipulate the beta parameter in the model, but it is not allowed. Any hint? Moreover, I tried to copy all the features of the model, but obviously the new "model" is actually a struct variable. Are there some functions to transform it in a model?

Thank you very much, Best Regards.