Support Vector Machine for binary classification only supports linear kernel, using SGD to optimize hinge loss update = lambda * (regularization(w) or marginal width or so called structure loss) + 1/batch * Hinge Loss bsz: default will be len(dataset)
(class_num: usize, feature_num: usize, lr: f32, bsz: Option<usize>, lambda: f32)
source not stored for this graph (policy: none)
nothing calls this directly
no outgoing calls
no test coverage detected