simple ngram for russian texts

ngram.py

import re
class NGram():
    pattern = re.compile('[^а-яА-Я ]+')
    _ngrams = dict()
    _it = 0
    
    def __init__(self):
        self.words = dict()
        self._it = 0

    def clean_text(self, s):
        return self.pattern.sub('', s)

    def find_ngrams(self, input_list, n):
        return zip(*[input_list[i:] for i in range(n)])
    
    def encode(self, string, n=3):
        clean_string = self.clean_text(string.lower())
        ngrams = self.find_ngrams(clean_string, n)
        codes = list()
        for ngram in ngrams:
            code = self._ngrams.get(ngram, None)
            if code is None:
                self._ngrams[ngram]=self._it
                code = self._it
                self._it += 1
            codes.append(code)
        return codes