Hands-On Python Deep Learning for the Web

Exploring the dataset

Let's begin by downloading all four files from the MNIST dataset web page, available at http://yann.lecun.com/exdb/mnist. Once downloaded, extract all the files and you should have folders that resemble the names in the following list:

• train-images.idx3-ubyte
• train-labels.idx1-ubyte
• t10k-images.idx3-ubyte
• t10k-labels.idx1-ubyte

Keep these files in your working directory. We will now create a Jupyter notebook to perform exploratory data analysis (EDA) on the dataset files we have extracted.

Open your Jupyter Notebook environment in your browser and create a new Python notebook. Let's begin by importing the necessary modules:

import numpy as npimport matplotlib.pyplot as plt

The preceding lines import the numpy module ...