Building a neural network for handwriting recognition

Now that we have loaded all that useful data, let's put it to good use. Since it's full of handwritten digits, we should most certainly build a model to recognize this handwriting and what it says.

In Chapter 2, What is a Neural Network and How Do I Train One?, we demonstrated how to build a simple neural network. Now, it's time to build something more substantial: a model for recognizing handwriting from the MNIST database.

Introduction to the model structure

First, let's think back to the original example: we had a single-layer network, which we wanted to get from a 4 x 3 matrix to a 4 x 1 vector. Now, we have to get from an MNIST image that is 28...