twenty_news_group.py 文件源码

def extract_feature(self):
        """
        ???????????
        """

        # ?????????-???
        self.train_dtm = self.count_vect.fit_transform(self.data['train'].data)

        # ????? TF ??

        tf_transformer = TfidfTransformer(use_idf=False)

        self.train_tf = tf_transformer.transform(self.train_dtm)

        # ????? TF-IDF ??

        tfidf_transformer = TfidfTransformer().fit(self.train_dtm)

        self.train_tfidf = tf_transformer.transform(self.train_dtm)