def _tf(self, word, count):
return count[word] / sum(count.values())
def _idf(self, word, count_list):
if word in self.idf_dict:
return self.idf_dict[word]
c_idf = sum(1 for q_id in count_list.keys() if word in count_list[q_id])
self.idf_dict[word] = math.log(len(count_list)) / (1 + c_idf)
return self.idf_dict[word]
def _tf_idf(self, word, count, count_list):
return self._tf(word, count) * self._idf(word, count_list)
Created
April 13, 2018 12:53
-
-
Save nlpjoe/88f985ea32e4a1fc378044dc22ec33e2 to your computer and use it in GitHub Desktop.
idf代码
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment