In this paper we will introduce main problems of supervised learning: regression and classification. Main part of the paper is divided into two sections, whereby first deals with regression problems, and the second deals with classification problems. We will show and explain the construction of mathematical models needed to solve these problems, point out the shortcomings of the basic models and show how to remove those shortcomings. At the end of the paper we will show the Python implementation of these models using scikit-learn module.