CNN basic example – MNIST digit classification

In this section, we will do a complete example of implementing a CNN for digit classification using the MNIST dataset. We will build a simple model of two convolution layers and fully connected layers.

Let's start off by importing the libraries that will be needed for this implementation:

%matplotlibinlineimportmatplotlib.pyplotaspltimporttensorflowastfimportnumpyasnpfromsklearn.metricsimportconfusion_matriximportmath

Next, we will use TensorFlow helper functions to download and preprocess the MNIST dataset as follows:

from tensorflow.examples.tutorials.mnist import input_data
mnist_data = input_data.read_data_sets('data/MNIST/', one_hot=True)

Output:
Successfully downloaded train-images-idx3-ubyte.gz 9912422 bytes.
Extracting data/MNIST/train-images-idx3-ubyte.gz
Successfully downloaded train-labels-idx1-ubyte.gz 28881 bytes.
Extracting data/MNIST/train-labels-idx1-ubyte.gz
Successfully downloaded t10k-images-idx3-ubyte.gz 1648877 bytes.
Extracting...