Python sklearn fit_transform()函数报错

Question

问题遇到的现象和发生背景

在文本主题建模的过程中出现了报错

问题相关代码：

for time in ['#B4','#C5','#C8','#D6','D23','D24']:
    
    raw_data=''
    for file in files:
        if re.match(str(time),file):
            with open(os.path.join(path,file), 'r',errors='ignore') as f:
                raw_data=raw_data+'#'+f.read()
                f.close()

    data=raw_data.split('#')

    data_vectorized = vectorizer.fit_transform(data)
 
    # Build a Latent Dirichlet Allocation Model
    lda_model = LatentDirichletAllocation(n_components=NUM_TOPICS, max_iter=10, learning_method='online')
    lda_Z = lda_model.fit_transform(data_vectorized)
 
    # Build a Non-Negative Matrix Factorization Model
    nmf_model = NMF(n_components=NUM_TOPICS)
    nmf_Z = nmf_model.fit_transform(data_vectorized)
 
    # Build a Latent Semantic Indexing Model
    lsi_model = TruncatedSVD(n_components=NUM_TOPICS)
    lsi_Z = lsi_model.fit_transform(data_vectorized)

    print_topics(lda_model, vectorizer)
    print_topics(nmf_model, vectorizer)
    print_topics(lsi_model, vectorizer)
    print(time)

运行结果及报错内容

#B4
#C5
#C8
#D6
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
/var/folders/85/h9zn7p197rv7q9qq734vvvsr0000gn/T/ipykernel_4497/1169374454.py in <module>
     10     data=raw_data.split('#')
     11 
---> 12     data_vectorized = vectorizer.fit_transform(data)
     13 
     14     # Build a Latent Dirichlet Allocation Model

~/opt/anaconda3/lib/python3.9/site-packages/sklearn/feature_extraction/text.py in fit_transform(self, raw_documents, y)
   1342             )
   1343             if max_doc_count < min_doc_count:
-> 1344                 raise ValueError("max_df corresponds to < documents than min_df")
   1345             if max_features is not None:
   1346                 X = self._sort_features(X, vocabulary)

ValueError: max_df corresponds to < documents than min_df

问题所在：

看不懂报错，不知道怎么解决

Answer 1

发现问题了，修改关键词"D23""D24"为“#D23”“#D24”，解决。