In this recipe, you will learn how to create a simple three-layer convolutional network to predict the MNIST digits. The deep network consists of two convolutional layers with ReLU and maxpool and two fully connected final layers.
Creating a ConvNet to classify handwritten MNIST numbers
Getting ready
MNIST is a set of 60,000 images representing handwritten numbers. The goal of this recipe is to recognize those numbers with high accuracy.
How to do it...
Let us start with the recipe:
- Import tensorflow, matplotlib, random, and numpy. Then, import the minst data and perform...