python - get_dummies 在 pandas 中的用法

Question

我正在阅读一本关于使用 Python 进行机器学习简介的书。这里作者描述如下假设对于 worker 类(Class)特征，我们可能的值为“政府”雇员”、“私有(private)雇员”、“个体运算符(operator)”和“个体运算符(operator)”特德”。

print("Original features:\n", list(data.columns), "\n")

data_dummies = pd.get_dummies(data)

print("Features after get_dummies:\n", list(data_dummies.columns))

Original features:
['age', 'workclass']

Features after get_dummies:
['age', 'workclass_ ?', 'workclass_ Government Employee', 'workclass_Private Employee', 'workclass_Self Employed', 'workclass_Self Employed Incorporated']

我的问题是新列 workclass_ 是什么？

Answer 1

它是使用workclass列的字符串值创建的:

data = pd.DataFrame({'age':[1,1,1,2,1,1],
                   'workclass':['Government Employee','Private Employee','Self Employed','Self Employed Incorpora ted','Self Employed Incorpora ted','?']})

print (data)
   age                    workclass
0    1          Government Employee
1    1             Private Employee
2    1                Self Employed
3    2  Self Employed Incorpora ted
4    1  Self Employed Incorpora ted
5    1                            ?

<小时/>

data_dummies = pd.get_dummies(data)
print (data_dummies)
   age  workclass_?  workclass_Government Employee  \
0    1            0                              1   
1    1            0                              0   
2    1            0                              0   
3    2            0                              0   
4    1            0                              0   
5    1            1                              0   

   workclass_Private Employee  workclass_Self Employed  \
0                           0                        0   
1                           1                        0   
2                           0                        1   
3                           0                        0   
4                           0                        0   
5                           0                        0   

   workclass_Self Employed Incorpora ted  
0                                      0  
1                                      0  
2                                      0  
3                                      1  
4                                      1  
5                                      0  

如果有多列具有相同的值，这个前缀非常有用:

data = pd.DataFrame({'age':[1,1,3],
                   'workclass':['Government Employee','Private Employee','?'],
                   'workclass1':['Government Employee','Private Employee','Self Employed']})

print (data)
   age            workclass           workclass1
0    1  Government Employee  Government Employee
1    1     Private Employee     Private Employee
2    3                    ?        Self Employed

data_dummies = pd.get_dummies(data)
print (data_dummies)
   age  workclass_?  workclass_Government Employee  \
0    1            0                              1   
1    1            0                              0   
2    3            1                              0   

   workclass_Private Employee  workclass1_Government Employee  \
0                           0                               1   
1                           1                               0   
2                           0                               0   

   workclass1_Private Employee  workclass1_Self Employed  
0                            0                         0  
1                            1                         0  
2                            0                         1  

如果不需要，可以添加参数，用空格覆盖:

data_dummies = pd.get_dummies(data, prefix='', prefix_sep='')
print (data_dummies)
   age  ?  Government Employee  Private Employee  Government Employee  \
0    1  0                    1                 0                    1   
1    1  0                    0                 1                    0   
2    3  1                    0                 0                    0   

   Private Employee  Self Employed  
0                 0              0  
1                 1              0  
2                 0              1  

然后可以按列进行groupby，并为每个唯一列的虚拟对象聚合max:

print (data_dummies.groupby(level=0, axis=1).max())
   ?  Government Employee  Private Employee  Self Employed  age
0  0                    1                 0              0    1
1  0                    0                 1              0    1
2  1                    0                 0              1    3