sklearn報錯Error message: fit_transform() takes 2 positional arguments but 3 were given的解決方法

最近在讀 Hands-On Machine Learning with Scikit-Learn & TensorFlow 這本書，在學到pipeline的時候，我模仿者寫了這也的代碼：

[python] view plain copy

1. num_attribs=list(housing_numerical)  
2. cat_attribs=["ocean_proximity"]  
3.   
4. num_pipeline=Pipeline([  
5.     ("selector",DataFrameSelector(num_attribs)),  
6.     ("imputer",Imputer(strategy="median")),  
7.     ("attribs_adder",CombinedAttributesAdder()),  
8.     ("std_scaler",StandardScaler()),  
9. ])  
10.   
11. cat_pipeline=Pipeline([  
12.     ("selector",DataFrameSelector(cat_attribs)),  
13.     'label_binarizer', LabelBinarizer()),
14. ])  
15.   
16. full_pipeline=FeatureUnion(transformer_list=[  
17.     ("num_pipeline",num_pipeline),  
18.     ("cat_pipeline",cat_pipeline),  
19. ])  

 

 

但是會報錯如下：

[plain] view plain copy

1. TypeError: fit_transform() takes 2 positional arguments but 3 were given  

我想，這應該是版本更新引起的問題，果然我在這裏找到了答案。以下爲引用：

 

The pipeline is assuming LabelBinarizer's fit_transform method is defined to take three positional arguments:

def fit_transform(self, x, y)
    ...rest of the code

while it is defined to take only two:

def fit_transform(self, x):
    ...rest of the code

所以，解決方法就是，自己寫一個根據LabelBinarizer寫一個MyLabelBinarizer,可以有三個參數self,X,y=None.

 

from sklearn.base import TransformerMixin #gives fit_transform method for free
class MyLabelBinarizer(TransformerMixin):
    def __init__(self, *args, **kwargs):
        self.encoder = LabelBinarizer(*args, **kwargs)
    def fit(self, x, y=0):
        self.encoder.fit(x)
        return self
    def transform(self, x, y=0):
        return self.encoder.transform(x)