如何解决“ AttributeError：使用LDA进行主题建模的模块'__main__'没有属性'__spec__'？”

Question

我正在为用于放射学报告的主题建模实施LDA。以下是我正在使用的代码

import pandas as pd
data = pd.read_csv('chest_ct.csv', error_bad_lines=False);

import gensim

def preprocess(text):
   result=[]
   for token in gensim.utils.simple_preprocess(text) :
      if token not in gensim.parsing.preprocessing.STOPWORDS and len(token) > 3:
         result.append((token))

   return result
processed_docs = []

for doc in data:
  processed_docs.append(preprocess(doc))
dictionary = gensim.corpora.Dictionary(processed_docs)
lda_model =  gensim.models.LdaMulticore(data, 
                               num_topics = 8, 
                               id2word = dictionary,                                    
                               passes = 2,
                               workers = 2)

该代码正在处理字典，并在lda_model处失败。我还尝试通过此链接https://pythonhosted.org/lda/使用scikit-learn实现LDA，并再次收到相同的错误。谁能帮忙吗？