This challenge involves taking MNIST images with noise, and trying to generate ones with the noise stripped away. 

The goal is to create a system which can successfully denoise it to the level that a human can understand it with the highest amount of noise possible. 

In order to generate the training and testing data, load in the MNIST dataset from keras and follow the instructions in the data file. 

Rules:
- All programs must be python3
- Use keras or tensorflow
- Submit your code by the end of the week