Python Deep Learning Projects

Chapter 2 – Training NN for Prediction Using Regression

In Chapter 2, Training NN for Prediction Using Regression, we identified our first business use case—one that would become a theme for a number of projects: that of a restaurant chain seeking to automate some of its processes. Specifically, in this chapter, the business use case was to build a deep learning classifier using a multi-layer perceptron (MLP), the basic building block in deep learning, to accurately classify handwritten digits of a customer's phone number. If you recall, the goal was to accurately classify (digitize) the handwritten phone number on an iPad so that the patron could receive a text that their table was ready.

We built a two-layer (minimally deep) neural network ...