Extending Scikit-Learn with your own regressor

Scikit-Learn is a well-known and popular framework for machine learning that is used by Data Scientists all over the world. In this tutorial presented at the EuroPython 2014 in Berlin, I show in a practical way how you can add your own estimator following the interfaces of Scikit-Learn. First a small introduction to the design of Scikit-Learn and its inner workings is given. Then I show how easily Scikit-Learn can be extended by creating an own estimator. In order to demonstrate this, I extend Scikit-Learn by the popular and robust Theil-Sen Estimator that was not in Scikit-Learn until version 0.16. I also motivate this estimator by outlining some of its superior properties compared to the ordinary least squares method (LinearRegression in Scikit-Learn).