继承自SciKit FunctionTransformer

Posted 2021-04-29

我想使用FunctionTransformer，同时提供一个简单的API并隐藏其他详细信息。具体来说，我希望能够提供一个Custom_Trans类，如下所示。因此，除了可以正常工作的trans1之外，用户还应该可以使用目前失败的trans2：

from sklearn import preprocessing 
from sklearn.pipeline import Pipeline
from sklearn import model_selection
from sklearn.linear_model import LinearRegression
from sklearn.datasets import make_regression
import numpy as np

X, y = make_regression(n_samples=100, n_features=1, noise=0.1)

def func(X, a, b):
    return X[:,a:b]

class Custom_Trans(preprocessing.FunctionTransformer):
    def __init__(self, ind0, ind1):
        super().__init__(
            func=func,
            kw_args={
                "a": ind0,
                "b": ind1
            }
        )

trans1 = preprocessing.FunctionTransformer( 
    func=func,
    kw_args={
        "a": 0,
        "b": 50
    }
)

trans2 = Custom_Trans(0,50)

pipe1 = Pipeline(
    steps=[
           ('custom', trans1),
           ('linear', LinearRegression())
         ]
)

pipe2 = Pipeline(
    steps=[
           ('custom', trans2),
           ('linear', LinearRegression())
          ]
)

print(model_selection.cross_val_score(
    pipe1, X, y, cv=3,)
)

print(model_selection.cross_val_score(
    pipe2, X, y, cv=3,)
)

这就是我得到的：

[0.99999331 0.99999671 0.99999772]
...sklearn/base.py:209: FutureWarning: From version 0.24, get_params will raise an
AttributeError if a parameter cannot be retrieved as an instance attribute. 
Previously it would return None.
warnings.warn('From version 0.24, get_params will raise an '
...
[0.99999331 0.99999671 0.99999772]

我有点知道这与估计器克隆有关，但我不知道如何解决。例如this post表示

估计器中应该没有逻辑，甚至没有输入验证init。逻辑应放在使用参数的位置，通常适合]

但是在这种情况下，我需要将参数传递给超类。无法将逻辑放入fit()中。我该怎么办？

答案

您可以通过从BaseEstimator继承来获取'get_params'。

class FunctionTransformer(BaseEstimator, TransformerMixin)

How to pass parameters to the customize modeltransformer class

inherit from function_transformer

custom transformers

您在基础中有此：

def get_params(self, deep=True):
        """
        Get parameters for this estimator.
        Parameters
        ----------
        deep : bool, default=True
            If True, will return the parameters for this estimator and
            contained subobjects that are estimators.
        Returns

更改您的代码：

trans1 = dict（functiontransformer__kw_args = [{'ind0'：无}，{'ind0'：1}]）

class Custom_Trans(preprocessing.FunctionTransformer): 
    def __init__(self, ind0, ind1, deep=True): 
        super().__init__( func=func, kw_args={ "a": ind0, "b": ind1 } ) 
        self.ind0 = ind0
        self.ind1 = ind1
        self.deep = True