Question

我正在创建垃圾邮件/火腿分类器。首先，我将所有电子邮件收录到矢量中。

然后，我使用sklearn的CountVectorizer对所有邮件中的单词计数，得到以下矩阵：

>> print(vector.shape)
>> print(type(vector))
>> print(vector.toarray())

(2551, 48746)
<class 'scipy.sparse.csr.csr_matrix'>
[[2 0 1 ... 0 0 0]
 [2 0 1 ... 0 0 0]
 [0 0 0 ... 0 0 0]
 ...
 [0 0 0 ... 0 0 0]
 [2 1 1 ... 0 0 0]
 [2 0 0 ... 0 0 0]]

如果尝试将向量更改为DataFrame，我得到：

>> df_X = pd.DataFrame(vector.toarray())

      0      1      2      3      4      5      6      7      8      ...  48737  48738  48739  48740  48741  48742  48743  48744  48745
0         2      0      1      0      0      0      0      0      0  ...      0      0      0      0      0      0      0      0      0
1         2      0      1      0      0      0      0      0      0  ...      0      0      0      0      0      0      0      0      0
2         0      0      0      0      0      0      0      0      0  ...      0      0      0      0      0      0      0      0      0
3         1      0      0      0      0      0      0      0      0  ...      4      0      0      0      0      0      0      0      0
4         3      0      1      0      0      0      0      0      0  ...      0      0      0      0      0      0      0      0      0
5

问题是我想为列指定有意义的名称（而不是0,1,2，...，48745）。

如果我运行print(vectorizer.vocabulary_)，我会得到：

>> print(vectorizer.vocabulary_)
{u'74282760403': 10172, u'makinglight': 34440, u'localizes': 33864, u'sowell': 43338, u'e4c8b2940d2': 22109, u'juob22381': 32587, u'31c6d68fa597d411b04d00e02965883bd239fb': 7072, u'20020918154734': 5469, u'spiders': 43495, u'ftrain': 24856, u'hanging': 30009, u'woody': 48041, u'000093': 18, u'1a724ef5': 4703, u'05dc347c66': 1771, u'g93ba2f21504': 28071, u'g16mteg13192': 25103, u'7f08f1c2c4': 10578, u'g974xhk18362': 28334, u'g85bc1j10899': 26181,...}

这是完整的代码：

import os,glob
import pandas as pd
from sklearn.feature_extraction.text import CountVectorizer 

folder_path = 'easy_ham/'
files_text_arr = []
files_text_arr_y = []

for filename in glob.glob(os.path.join(folder_path, '*')):
  with open(filename, 'r') as f:
    text = f.read()
    files_text_arr.append(text)
    files_text_arr_y.append(0)

vectorizer = CountVectorizer(encoding='latin-1')
vectorizer.fit(files_text_arr)

vector = vectorizer.transform(files_text_arr)
print(vector.shape)
print(type(vector))
print(vector.toarray())
#print(vectorizer.vocabulary_)    

df_X = pd.DataFrame(vector.toarray())
df_y = pd.DataFrame({'spam':files_text_arr_y})
print(df_X)

如何将列的名称更改为电子邮件中的单词？

P.S。我使用来自this website的电子邮件。

Answer 1

您可以使用方法get_feature_names()，然后将其分配给由toarray()方法的输出创建的数据框的列。

from sklearn.feature_extraction.text import CountVectorizer
corpus = [
    'This is the first document.',
    'This document is the second document.',
    'And this is the third one.',
    'Is this the first document?',
]
vectorizer = CountVectorizer()
X = vectorizer.fit_transform(corpus)
print(vectorizer.get_feature_names())

print(X.toarray())

输出

[u'and', u'document', u'first', u'is', u'one', u'second', u'the', u'third', u'this']
[[0 1 1 1 0 0 1 0 1]
 [0 2 0 1 0 1 1 0 1]
 [1 0 0 1 1 0 1 1 1]
 [0 1 1 1 0 0 1 0 1]]

df = pd.DataFrame(X.toarray())
df.columns = vectorizer.get_feature_names()
df

输出

    and document    first   is  one second  the third   this
 0   0      1        1      1    0    0       1    0    1
 1   0      2        0      1    0    1       1    0    1
 2   1      0        0      1    1    0       1    1    1
 3   0      1        1      1    0    0       1    0    1

命名从CountVectorizer Vector创建的DataFrame的列

1 个答案: