具有多个输入/输出的sklearn管道

时间:2019-09-20 12:04:37

标签: scikit-learn pytorch

如何建立sklearn管道来执行以下操作?

我所拥有的:

A, B = getAB(X_train)
X_train = transform(X_train)
model(A, B, X_train)

我想要什么:

pipe = Pipeline([
(‘ab’, getAB),
(‘tranf’, transform),
(‘net’, net)
]
pipe.fit(X_train, y_train)

请帮助!

1 个答案:

答案 0 :(得分:1)

是的,可以通过编写具有拟合/转换功能的自定义转换器来实现。这可以是您的课程:

from sklearn.base import BaseEstimator, TransformerMixin

def getABTransformer(BaseEstimator, TransformerMixin):
    def __init__(self): # no *args or **kargs
        pass

    def fit(self, X, y=None):
        return self # nothing else to do

    def transform(self, X, y=None):
        return getAB(X)

然后,您可以按照以下步骤创建ColumnTransformer

from sklearn.compose import ColumnTransformer

clm_pipe = ColumnTransformer([
(‘ab’, getABTransformer, np.arange(0, len(X_train)),  # list of columns indices
(‘tranf’, transform, np.arange(0, len(X_train))),  # list of columns indices
]

以及该模型的最终管道:

pipe = Pipeline([
(‘clm_pipe’, clm_pipe),
(‘net’, net)
]

您可以阅读有关ColumnTransformer

的更多信息