Function ngrams

pattern/metrics.py:341–351

Returns a list of n-grams (tuples of n successive words) from the given string. Punctuation marks are stripped from words.

(string, n=3, punctuation=PUNCTUATION, **kwargs)

PUNCTUATION = ".,;:!?()[]{}`''\"@#$^&*+-|=~_"

def ngrams(string, n=3, punctuation=PUNCTUATION, **kwargs):
    """ Returns a list of n-grams (tuples of n successive words) from the given string.
        Punctuation marks are stripped from words.
    """
    s = string
    s = s.replace(".", " .")
    s = s.replace("?", " ?")
    s = s.replace("!", " !")
    s = [w.strip(punctuation) for w in s.split()]
    s = [w.strip() for w in s if w.strip()]
    return [tuple(s[i:i+n]) for i in range(len(s)-n+1)]

class Weight(float):
    """ A float with a magic "assessments" property,