Abstract: AutoML and Auto-Sklearn

by Matthias Feurer and Katharina Eggensperger

In this hands-on, we will cover how to move from manually constructing and tuning machine learning pipelines developed with scikit-learn to using efficient hyperparameter optimization algorithms and full AutoML using the popular open-source Auto-sklearn library, which is a drop-in replacement for any scikit-learn estimator. We will go through the various settings of the library, demonstrate how it can be used parallel settings and apply it to various example datasets.