Uses the MNIST dataset with 60,000 images of handwritten digits and 10,000 testing images. Goal of the project was to create a Naive Bayes classifier to match the images to their respective labels. After training I created three confusion matrices using the Seaborn package, these are included in the report PDF HERE.