This example (like all the others in this and the following chapters) is based on TensorFlow (for information about the installation of TensorFlow, please refer to the information box at the end of the section), because this framework allows a greater flexibility that is sometimes much more problematic with Keras. We will approach this example pragmatically, and so we are not going to explore all the features because they are beyond the scope of this book; however, interested readers can refer to Deep Learning with TensorFlow - Second Edition, Zaccone G., Karim R., Packt.
In this example, we are going to create a deep convolutional autoencoder and train it using the Fashion MNIST ...
