MNIST Classification

Digit Classification using a Convolutional Neural Network for Image Processing (Tensorflow 1.12.0 and Python 3.6.6)

• Editor used: Sublime Text 3
• Shell used to run the code: Git Bash
• Libraries used: Numpy, Matplotlib, PIL & Tensorflow
• The purpose of the model is to be able to distinguish between digits (0-9) and ultimately, be able to determine which digit is being shown to the model during the prediction step
• sample_image.jpg is the image which is shown to the model during the prediction step

Insight on the MNIST database

• The MNIST database contains 60,000 training images and 10,000 testing images
• Each image has a size of 28x28 pixels and consists of a single channel