So I have a dataset with symbols I\'d like to clean up using df.replace:
and replace them with an apostrophe (most of the