Clean dataset

[dahkimr]

Code to remove unnecessary punction, etc.

Resource: https://towardsdatascience.com/nlp-in-python-data-cleaning-6313a404a470

• decide which datasets to run methods on
• run code on combined dataset

[Squishy123]

We can also look to combine datasets into multiple csvs to make it easier I think having 3 training datasets and 3 testing datasets would be optimal

Something like a small, medium and large dataset.