python one-hot符号化を実現

6738 ワード

''' one-hot  '''
from sklearn.preprocessing import OneHotEncoder 
from sklearn.preprocessing import LabelEncoder
from sklearn.preprocessing import LabelBinarizer
from sklearn.preprocessing import MultiLabelBinarizer

testdata = pd.DataFrame({'pet': ['chinese', 'english', 'english', 'math'],
                         'age': [6 , 5, 2, 2],
                         'salary':[7, 5, 2, 5]})

''' OneHotEncoder  '''
 result:
 >>> testdata
   age      pet  salary
0    6  chinese       7
1    5  english       5
2    2  english       2
3    2     math       5


OneHotEncoder(sparse = False).fit_transform(testdata[['age']])
result;
>>> OneHotEncoder(sparse = False).fit_transform(testdata[['age']])
array([[ 0.,  0.,  1.],
       [ 0.,  1.,  0.],
       [ 1.,  0.,  0.],
       [ 1.,  0.,  0.]])

 ''' one-hot '''
 >>> OneHotEncoder(sparse = False).fit_transform( testdata[['age', 'salary']])
array([[ 0.,  0.,  1.,  0.,  0.,  1.],
       [ 0.,  1.,  0.,  0.,  1.,  0.],
       [ 1.,  0.,  0.,  1.,  0.,  0.],
       [ 1.,  0.,  0.,  0.,  1.,  0.]])


'''  one-hot  '''
>>> LabelEncoder().fit_transform(testdata['pet'])
array([0, 1, 1, 2], dtype=int64)
>>> testdata
   age      pet  salary
0    6  chinese       7
1    5  english       5
2    2  english       2
3    2     math       5
>>> LabelEncoder().fit_transform(testdata['pet'])
array([0, 1, 1, 2], dtype=int64)
>>> a = LabelEncoder().fit_transform(testdata['pet'])
>>> OneHotEncoder( sparse=False ).fit_transform(a.reshape(-1,1))
array([[ 1.,  0.,  0.],
       [ 0.,  1.,  0.],
       [ 0.,  1.,  0.],
       [ 0.,  0.,  1.]])

###########   ############

>>> LabelBinarizer().fit_transform(testdata['pet'])
array([[1, 0, 0],
       [0, 1, 0],
       [0, 1, 0],
       [0, 0, 1]])