This notebook uses deep learning with fast AI and grabs the MNIST dataset from Kaggle. It then trains and tests a model using fast ai and creates a csv file for submission to Kaggle.
Most of the foundation of the code was from lesson 2 of fastai's Practical Deep Learning for Coders found here https://course.fast.ai/.
The final submission to Kaggle got a public score of 0.97396.
Resources that contributed code to this notebook:
https://www.kaggle.com/stefanbuenten/mnist-test-with-fastai
https://www.kaggle.com/daneiref/yet-another-mnist-using-fastai