使用pandas填写缺失的分类值？

Question

我想用每列新值填充缺失的分类单元格。例如：

c1  c2  c3
a   nan  a
b   q    nan
c   d    nan
a   p    z

应该变得像

c1  c2  c3
a   n1   a
b   q    n2
c   d    n2
a   p    z

我目前的问题是我使用DictVectorizer作为分类列，但它保留了原样的NaN。

Answer 1

带有一些uniq字符串的Fillna可以满足您的需求：

categorial_data = pd.DataFrame({'sex': ['male', 'female', 'male', 'female'],
                                'nationality': ['American', 'European', float('nan'), 'European']})
print(categorial_data)
categorial_data=categorial_data.fillna('some_unique_string')
print('after replacement')
print(categorial_data)
encoder = DV(sparse = False)
encoded_data = encoder.fit_transform(categorial_data.T.to_dict().values())
print(encoded_data)

给你

  nationality     sex
0    American    male
1    European  female
2         NaN    male
3    European  female
after replacement
          nationality     sex
0            American    male
1            European  female
2  some_unique_string    male
3            European  female
[[ 1.  0.  0.  0.  1.]
 [ 0.  1.  0.  1.  0.]
 [ 0.  0.  1.  0.  1.]
 [ 0.  1.  0.  1.  0.]]