Extracts top of bigrams (ngrams) from text.

bigram_extractor.py

#!/usr/bin/env python3
from collections import Counter
from re import findall
from sys import argv


def main(text, top, n=2):
    ngrams = []
    for word in findall(r'\w+', text.lower()):
        wlen = len(word)
        if wlen >= n:
            ngrams.extend({word[i:i + n] for i in range(wlen - n + 1)})
    print([k for k, _ in Counter(ngrams).most_common(top)])


if __name__ == '__main__':
    main(argv[1], argv[2] if len(argv) == 3 else 300)