Question

我尝试在实例化类后调用_get_tfidf应用程序。考虑以下脚本：

text2vec.py

class text2vec():
    def __init__(self, doc_list):
        #Initialize
        self.doc_list = doc_list
        self.nlp, self.docs, self.docs_dict = self._preprocess(self.doc_list)

    def get_tfidf(self):
        docs_corpus = [self.docs_dict.doc2bow(doc) for doc in self.docs]
        model_tfidf = TfidfModel(docs_corpus, id2word=self.docs_dict)
        docs_tfidf  = model_tfidf[docs_corpus]
        docs_vecs   = np.vstack([sparse2full(c, len(self.docs_dict)) for c in docs_tfidf])
        return docs_vecs

    def _get_docs_dict(self, docs):
        docs_dict = Dictionary(docs)
        #CAREFUL: For small corpus please carefully modify the parameters for filter_extremes, or simply comment it out.
        docs_dict.filter_extremes(no_below=5, no_above=0.2)
        docs_dict.compactify()
        return docs_dict

    # Preprocess docs
    def _preprocess(self, doc_list):
        #Load spacy model
        nlp  = spacy.load('en_core_web_sm')
        #lemmatise docs
        docs = [self._lemmatize_doc(nlp(doc)) for doc in doc_list] 
        #Get docs dictionary
        docs_dict = self._get_docs_dict(docs)
        return nlp, docs, docs_dict

main.py

import text2vec

t2v = text2vec.text2vec(doc_list)

# Use TFIDF
docs_tfidf = t2v.get_tfidf()

但是我得到这个错误：

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-23-04f280ed20cf> in <module>()
      1 # Use TFIDF
----> 2 docs_tfidf = t2v.get_tfidf()

~/SageMaker/sentences matching/text2vec.py in get_tfidf(self)
     84     # Get TF-IDF vector for document list
     85     def get_tfidf(self):
---> 86         docs_corpus = [self.docs_dict.doc2bow(doc) for doc in self.docs]
     87         model_tfidf = TfidfModel(docs_corpus, id2word=self.docs_dict)
     88         docs_tfidf  = model_tfidf[docs_corpus]

AttributeError: 'text2vec' object has no attribute 'docs'

有什么想法吗？

谢谢

Python没有类的属性

0 个答案: