What is the best machine learning algorithm that can be used to calculate exactly its decision boundaries for a 2D or ND classification for numerical data. I’m not interested in the performance of the classification but the simplicity and the calculation time of the decision boundaries. I have already used SVM with sk-learn but it returns just the equation of each line but not the polygons delimiting the classes in 2D. I’m interested in extracting the polygons or the exact shape of the final classification for each class.
Thank you so much.

I tried SVM but the library doesn’t provide the solution. I’m expecting a supervised machine learning algorithm with easy to extract decision boundaries method.