yes, I handled that case as you described it

then based on this sample test:

Test.assert_equals(top_3_words(" //wont won't won't "), ["won't", "wont"])

are we not required to filter out characters that are from A-Z or (')?

EDIT: this happens using python 3.6
some of the random test cases test numbers when they were not mentioned in the instructions.

A word is a string of letters (A to Z) optionally containing one or more apostrophes (') in ASCII. (No need to handle fancy punctuation.)

 Frequencies of the most common words: [13, 25, 19] should equal [25, 25, 23]

also, on the same note: The random test expect duplicate results