Question

我正在构建一个情感分析算法，可以对.txt语料库中的分段进行分段，但代码中存在一些问题，我不知道如何解决？

class Splitter(object):
def _init_(self):
    self.nltk_splitter = nltk.data.load('tokenizers/punkt/english/pickle')
    self.nltk_tokenizer = nltk.tokenize.TreebankWordTokenizer()
def split(self,text):
    """imput format: a .txt file
               output format : a list of lists of words.
               for eg [['this', 'is']['life' , 'worth' , 'living']]"""
    sentences = self.nltk_splitter.tokenize(text)
    tokenized_sentences = [self.nltk_tokenizer.tokenize(sent) for sent in sentences]
    return tokenized_sentences

然后我做了以下事情

 >>> f = open('amazonshoes.txt')
 >>> raw = f.read()
 >>> text = nltk.Text(raw)
 >>> splitter = Splitter()
 >>> splitted_sentences = splitter.split(text)

，错误是

  Traceback (most recent call last):
    File "<pyshell#21>", line 1, in <module>
      splitted_sentences = splitter.split(text)
    File "<pyshell#14>", line 9, in split
      sentences = self.nltk_splitter.tokenize(text)
AttributeError: 'Splitter' object has no attribute 'nltk_splitter'

Answer 1

类Splitter的构造函数应该被称为__init__，带有两个前导和尾部下划线。

目前未执行_init_方法（单个下划线），因此您创建的Splitter对象（通过调用Splitter()）永远不会获取属性/字段nltk_splitter < / p>

python，nltk 2.7在代码中显示错误，可能是什么解决方案？

1 个答案: