scikit learn:ValueError:np.nan是无效的文档

Question

我正在使用scikit中的TfidfVectorizer学习从文本数据中提取一些特征。我有一个CSV文件，有一个分数（可以是+1或-1）和一个评论（文本）。我把这些数据放到一个数据框中，这样我就可以运行矢量器了。

这是我的代码：

import pandas as pd
import numpy as np
from sklearn.feature_extraction.text import TfidfVectorizer

df = pd.read_csv("train_new.csv",
             names = ['Score', 'Review'], sep=',')

# x = df['Review'] == np.nan
#
# print x.to_csv(path='FindNaN.csv', sep=',', na_rep = 'string', index=True)
#
# print df.isnull().values.any()

v = TfidfVectorizer(decode_error='replace', encoding='utf-8')
x = v.fit_transform(df['Review'])

这是我得到的错误的回溯：

Traceback (most recent call last):
  File "/home/PycharmProjects/Review/src/feature_extraction.py", line 16, in <module>
x = v.fit_transform(df['Review'])
 File "/home/b/hw1/local/lib/python2.7/site-   packages/sklearn/feature_extraction/text.py", line 1305, in fit_transform
   X = super(TfidfVectorizer, self).fit_transform(raw_documents)
 File "/home/b/work/local/lib/python2.7/site-packages/sklearn/feature_extraction/text.py", line 817, in fit_transform
self.fixed_vocabulary_)
 File "/home/b/work/local/lib/python2.7/site- packages/sklearn/feature_extraction/text.py", line 752, in _count_vocab
   for feature in analyze(doc):
 File "/home/b/work/local/lib/python2.7/site-packages/sklearn/feature_extraction/text.py", line 238, in <lambda>
tokenize(preprocess(self.decode(doc))), stop_words)
 File "/home/b/work/local/lib/python2.7/site-packages/sklearn/feature_extraction/text.py", line 118, in decode
 raise ValueError("np.nan is an invalid document, expected byte or "
 ValueError: np.nan is an invalid document, expected byte or unicode string.

我检查了CSV文件和DataFrame，看是否有被读取为NaN的内容，但找不到任何内容。共有18000行，没有一行返回isnan为True。

这就是df['Review'].head()的样子：

  0    This book is such a life saver.  It has been s...
  1    I bought this a few times for my older son and...
  2    This is great for basics, but I wish the space...
  3    This book is perfect!  I'm a first time new mo...
  4    During your postpartum stay at the hospital th...
  Name: Review, dtype: object

Answer 1

我找到了一个更有效的方法来解决这个问题。

x = v.fit_transform(df['Review'].apply(lambda x: np.str_(x)))

当然，您可以使用df['Review'].values.astype('U')来转换整个系列。但是我发现如果要转换的序列非常大，使用这个函数将消耗更多的内存。（我用一个80w行数据的系列测试这个，这样做astype('U')将消耗大约96GB的内存）

相反，如果使用lambda表达式只将序列中的数据从str转换为numpy.str_，结果也将被fit_transform函数接受，则速度会更快，不会增加内存使用量。

我不确定为什么这样做，因为在TFIDF矢量器的Doc页中：

fit_transform(raw_documents, y=None)

Parameters: raw_documents : iterable

an iterable which yields either str, unicode or file objects

但实际上这个iterable必须产生np.str_，而不是str。

Answer 2

您需要将数据类型object转换为unicode字符串，正如回溯中明确提到的那样。

x = v.fit_transform(df['Review'].values.astype('U'))  ## Even astype(str) would work

从TFIDF矢量器的Doc页：

fit_transform(raw_documents, y=None)

Parameters: raw_documents : iterable
an iterable which yields either str, unicode or file objects