Python 使用n_gram的文本摘要（如何编写函数对句子进行评分）

我有如下生成n_gram的代码：

import nltk
from nltk.util import ngrams

def extract_ngrams(data, num):
    n_grams = ngrams(nltk.word_tokenize(data.lower()),num)
    return [' '.join(grams) for grams in n_grams]

gram_num = 3

n_grams = extract_ngrams(filtered_sentence, gram_num)

from nltk import FreqDist
ngram_freqs = FreqDist(n_grams)

sent_tokens = nltk.sent_tokenize(text)

我使用fdist获得了n-gram频率的代码，如下所示：

import nltk
from nltk.util import ngrams

def extract_ngrams(data, num):
    n_grams = ngrams(nltk.word_tokenize(data.lower()),num)
    return [' '.join(grams) for grams in n_grams]

gram_num = 3

n_grams = extract_ngrams(filtered_sentence, gram_num)

from nltk import FreqDist
ngram_freqs = FreqDist(n_grams)

sent_tokens = nltk.sent_tokenize(text)

但我不知道如何定义给句子打分的函数 我想得到如下函数：

calculate_sentence_scores_ngram(sent_tokens, ngram_freqs, n_grams)