In this chapter, we learned about Keras. Keras is the most popular high-level library for TensorFlow. I personally prefer to use Keras for all the models that I develop for my commercial production work and also for academic research. We learned the workflow that we can follow to create and train the models in Keras, using both the functional and sequential APIs. We learned about various Keras layers and how to add the layers to the sequential and functional models. We also learned how to compile, train, and evaluate the Keras models. We also saw some of the additional modules provided by Keras.
Throughout the remaining chapters of the book, we shall be covering most of the examples in both core TensorFlow and Keras. In the next chapter, we will learn how to use TensorFlow for building traditional machine learning models for classification and regression.
...
