import numpy as np def split_by_words(X): X = np.core.chararray.lower(X) return np.core.chararray.split(X) DELIMITERS = "!?:;,.\'-+/\\()" def parse(string): return "".join((" " if char in DELIMITERS else char) for char in string).split()