SVM from scratch in Julia

Repo including explanation and high-level derivation of the SVM method used for classification, see github.shilangyu.dev/SVM-from-scratch. Additionally implementations of 3 different SVMs models can be found in Julia files.

Linear soft margin SVM

Non linear RBF kernel SVM

training does not look continuous because the optimization method takes harsh steps which converge quickly