python - CountVectorizer 中的样本数量不一致

Question

我正在尝试对我拥有的一组推文使用多项式朴素贝叶斯分类。

这是我的代码:

import codecs
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.naive_bayes import MultinomialNB
trainfile = 'train.txt'
testfile = 'test.txt'
word_vectorizer = CountVectorizer(analyzer='word')
trainset = word_vectorizer.fit_transform(codecs.open(trainfile,'r','utf8')) ## Error here
tags = ['Pro_vax','Anti_vax','Neither']
mnb = MultinomialNB()
mnb.fit(trainset, tags)
codecs.open(testfile,'r','utf8')
testset = word_vectorizer.transform(codecs.open(testfile,'r','utf8'))
results = mnb.predict(testset)
print results

文件 train.txt 中包含以下文本:

Vaccines are a very good idea.  They prevent all sorts of deadly diseases.
Vaccines cause autism.  Do not vaccinate your children
Going to read about vaccines.  Then, I am going to see my brother with autism.

我已使用 tags 变量对它们进行了标记。

文件test.txt具有以下文本:

Do not get your kids vaccinated.  Vaccination and autism are correlated.

当我运行脚本时，出现以下错误:

ValueError: Found arrays with inconsistent numbers of samples: [3 9]

我不熟悉该错误。这是什么意思？如何防止它再次弹出？

Answer 1

如果提供完整的回溯，会更容易看出，但看起来标签包含 9 个条目，而 train 仅包含三个训练数据点。 标签是什么样的？