Handwritten digits recognition

The challenge of handwritten digit recognition is to recognize digits from images of handwritten digits. It is useful in many scenarios, for example recognizing zip codes on envelopes. In this example, we will use the MNIST dataset to develop and evaluate our neural network model for handwritten digit recognition.

MNIST is a computer vision dataset hosted at: http://yann.lecun.com/exdb/mnist/. It consists of grayscale images of handwritten digits along with the correct digit labels. Each image is 28 pixels by 28 pixels. Sample images are shown as follows:

Sample images from MNIST dataset

The MNIST data is split ...

Get Deep Learning Essentials now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.