9 Wrapping up

 

This chapter covers

  • Important takeaways from this book
  • Open source toolkits and commercial platforms for AutoML
  • The challenges and future of AutoML
  • Resources for learning more and working in the field

We’ve almost reached the end of the book. This last chapter reviews the core concepts we’ve covered, while also aiming to expand your horizons. We’ll start with a quick recap of what you should take away from this book. Next, we’ll present an overview of some popular AutoML tools (both open source and commercial) outside the Keras ecosystem. An awareness of other emblematic toolkits in the current AutoML community will enable you to explore further based on your interests after reading the book. Finally, we offer some speculative thoughts about the core challenges and future evolution in the AutoML domain, which will be of particular interest if you’d like to delve into more fundamental research in this area. Understanding AutoML is a journey, and finishing this book is merely the first step. At the end of the chapter, we’ll provide you with a short list of resources and strategies for learning more about AutoML and staying up-to-date with the latest developments in the field.

9.1 Key concepts in review

This section briefly summarizes the key takeaways from this book, to refresh your memory of what you’ve learned so far.

9.1.1 The AutoML process and its key components

9.1.2 The machine learning pipeline

9.1.3 The taxonomy of AutoML

9.1.4 Applications of AutoML

9.1.5 Automated deep learning with AutoKeras

9.1.6 Fully personalized AutoML with KerasTuner

9.1.7 Implementing search techniques

9.1.8 Scaling up the AutoML process

9.2 AutoML tools and platforms

9.2.1 Open source AutoML tools

9.2.2 Commercial AutoML platforms

9.3 The challenges and future of AutoML

9.3.1 Measuring the performance of AutoML