Classifying handwritten digits

In the previous section, we covered a lot of the theory around neural networks, which can be a little overwhelming if you are new to this topic. In this section, we will classify handwritten digits from the popular MNIST dataset, which has been constructed by Yann LeCun and colleagues and serves as a popular benchmark dataset for machine learning algorithms.

We will train two different networks on it:

• a multilayer perceptron using OpenCV
• a deep neural net using Keras

Loading the MNIST dataset

The easiest way to obtain the MNIST dataset is using Keras:

In [1]: from keras.datasets import mnist
...     (X_train, y_train), (X_test, y_test) = mnist.load_data()
Out[1]: Using TensorFlow backend.
        Downloading data from
        https://s3.amazonaws.com/img-datasets/mnist.npz

This will download the data from the Amazon Cloud (might take a while depending on your internet connection) and automatically split the data into training and test sets.

Note

MNIST provides its own...