Neural network is a supervised learning algorithm that is loosely inspired by the way the brain functions. Similarly to the way neurons are connected to each other in the brain, a neural network takes an input and passes it through a function, based on which certain subsequent neurons get excited, and the output is produced.
In this chapter, we will focus on the practical implementation of neural networks with TensorFlow and Keras. TensorFlow provides a low-level framework to create neural network models. Keras is a high-level neural network API that significantly simplifies the task of defining neural network models. We'll show how to use Keras on top of TensorFlow to define and train models on GCP. We'll present the Keras API in Python and work with a simple feedforward network applied on the classic MNIST dataset. Also, we...
