AttributeError：“tokenizer”对象没有“”sent_tokenize“”属性“”
AttributeError：“tokenizer”对象没有“”sent_tokenize“”属性“”
提问于 2018-06-04 16:45:58
当我运行下面的代码时，我得到了以下错误：sentence_token= self.sent_tokenize(input_data)行上的AttributeError: 'tokenizer' object has no attribute 'sent_tokenize'。
为什么它会给我这个错误？
from nltk.tokenize import sent_tokenize, \word_tokenize, WordPunctTokenizer
from nltk.corpus import brown
class tokenizer:
input_data=''
def __init__(self, input_data='', ):
self.input_data = input_data
def gen_SentTokenizers(self,input_data):
sentence_token= self.sent_tokenize(input_data) # Error!
return sentence_token
def gen_WordTokenizers(self, input_data):
word_token=self.word_tokenize(input_data)
return word_token
def gen_WordPunctToken(self, input_data):
word_PunctToken=self.WordPunctTokenizer().tokenize(input_data)
return word_PunctToken
if __name__=='__main__':
input_data = ' '.join(brown.words())
token1 = tokenizer()
token2 = tokenizer()
token3 = tokenizer()
token1.gen_SentTokenizers(input_data)
print("\nSentence tokenizer:")
print(token1)
token2.gen_WordPunctToken(input_data)
print("\nWord punct tokenizer:")
print(token2)
token3.gen_WordTokenizers(input_data)
print("\nWord tokenizer:")
print(token3) 回答 1
Stack Overflow用户
发布于 2018-06-04 16:56:22
在终端上：
pip install --upgrade nltk
python -m nltk.downloader popular在代码中：
from nltk.tokenize import sent_tokenize
text = 'This str has many sentences. A sentence here. A sentence there.'
sent_tokenize(text)页面原文内容由Stack Overflow提供。腾讯云小微IT领域专用引擎提供翻译支持
原文链接：
https://stackoverflow.com/questions/50676612
复制相关文章
相似问题