This open access book provides a wealth of hands-on examples thatillustrate how hyperparameter tuning can be applied in practice andgives deep insights into the working mechanisms of machine learning (ML)and deep learning (DL) methods. The aim of the book is to equip readerswith the ability to achieve better results with significantly lesstime, costs, effort and resources using the methods described here. Thecase studies presented in this book can be run on a regular desktop ornotebook computer. No high-performance computing facilities arerequired. The idea for the book originated in a study conducted by Bartz& Bartz GmbH for the Federal Statistical Office of Germany(Destatis). Building on that study, the book is addressed topractitioners in industry as well as researchers, teachers and studentsin academia. The content focuses on the hyperparameter tuning of ML andDL algorithms, and is divided into two main parts: theory (Part I) andapplication (Part II). Essential topics covered include: a survey ofimportant model parameters; four parameter tuning studies and oneextensive global parameter tuning study; statistical analysis of theperformance of ML and DL methods based on severity; and a new,consensus-ranking-based way to aggregate and analyze results frommultiple algorithms. The book presents analyses of more than 30hyperparameters from six relevant ML and DL methods, and provides sourcecode so that users can reproduce the results. Accordingly, it serves asa handbook and textbook alike.