corpus.py 文件源码

def sents(self, fileids=None, categories=None):
        """
        Uses the built in sentence tokenizer to extract sentences from the
        paragraphs. Note that this method uses BeautifulSoup to parse HTML.
        """
        for paragraph in self.paras(fileids, categories):
            for sentence in self._sent_tokenizer.tokenize(paragraph):
                yield sentence