ValueError: max_df corresponds to < documents than min_df

Question

问题遇到的现象和发生背景

在跑LDA模型的时候报错，应该是在tf-idf向量化的时候报错的。

问题相关代码

n_features = 1000 #提取1000个特征词语
tf_vectorizer = CountVectorizer(strip_accents = 'unicode',
                                max_features=n_features,
                                stop_words='english',
                                max_df = 0.5,
                                min_df = 10)
tf = tf_vectorizer.fit_transform(data.content_cutted)

报错内容

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-11-ee1a3704afca> in <module>
      5                                 max_df = 0.5,
      6                                 min_df = 10)
----> 7 tf = tf_vectorizer.fit_transform(data.content_cutted)

D:\anaconda3\lib\site-packages\sklearn\feature_extraction\text.py in fit_transform(self, raw_documents, y)
   1216                              else min_df * n_doc)
   1217             if max_doc_count < min_doc_count:
-> 1218                 raise ValueError(
   1219                     "max_df corresponds to < documents than min_df")
   1220             if max_features is not None:

ValueError: max_df corresponds to < documents than min_df

Answer 1

我猜应该是样本量少了，max和min之间的区间相对就太大了，我把min_df改成2就能运行了