Keras 101

Keras is a high-level library that allows the use of TensorFlow as a backend deep learning library. TensorFlow team has included Keras in TensorFlow Core as module tf.keras. Apart from TensorFlow, Keras also supports Theano and CNTK at the time of writing this book.

The following guiding principles of Keras have made it very popular among the deep learning community:

• Minimalism to offer a consistent and simple API
• Modularity to allow the representation of various elements as pluggable modules
• Extensibility to add new modules as classes and functions
• Python-native for both code and model configuration
• Out-of-the-box common network architectures that support CNN, RNN, or a combination of both

Throughout the remainder of this book, we shall learn how to build different kinds of deep learning and machine learning models with both the low-level TensorFlow API and the high-level Keras API.

We shall cover the following topics in this chapter:

• Installing Keras
• Workflow for creating models in Keras
• Creating the Keras model with sequential and functional API
• Keras layers
• Creating and adding layers with sequential and functional API

• Compiling the Keras model
• Training the Keras model
• Predicting using the Keras model
• Additional modules in Keras
• Keras sequential model example for MNIST dataset