如何在 scikit 学习列选择器管道中只选择几列？

Question

我正在阅读有关列转换器的 scikitlearn 教程。给定的示例（https://scikit-learn.org/stable/modules/generated/sklearn.compose.make_column_selector.html#sklearn.compose.make_column_selector）有效，但是当我尝试仅选择几列时，它给了我错误.

移动电源

import numpy as np
import pandas as pd
import seaborn as sns

from sklearn.compose import make_column_transformer
from sklearn.compose import make_column_selector

df = sns.load_dataset('tips')
mycols = ['tip','sex']


ct = make_column_transformer(make_column_selector(pattern=mycols)
ct.fit_transform(df)

必需的

我只想要输出中的选择列。

注意
当然，我知道我可以做到df[mycols]，我正在寻找 scikit 学习管道示例。

Answer 1

ColumnTranformer()我可能有点晚了，但您也可以通过将变压器设置为“passthrough”来使用 sklearn 选择列remainder='drop'：

from sklearn.compose import ColumnTransformer
from sklearn.pipeline import Pipeline


pipe = Pipeline([
    ("selector", ColumnTransformer([
        ("selector", "passthrough", mycols)
    ], remainder="drop"))
])

Answer 2

如果你不介意mlxtend，它有内置的变压器。

使用 mlxtend

from mlxtend.feature_selection import ColumnSelector

pipe = ColumnSelector(mycols)
pipe.fit_transform(df)

对于 sklearn >= 0.20

• 参考：https : //scikit-learn.org/stable/modules/generated/sklearn.compose.ColumnTransformer.html

from mlxtend.feature_selection import ColumnSelector

pipe = ColumnSelector(mycols)
pipe.fit_transform(df)

对于 sklearn < 0.20

from sklearn.base import BaseEstimator, TransformerMixin
from sklearn.pipeline import Pipeline

class FeatureSelector(BaseEstimator, TransformerMixin):
    def __init__(self, columns):
        self.columns = columns

    def fit(self, X, y=None):
        return self

    def transform(self, X, y=None):
        return X[self.columns]


pipeline = Pipeline([('selector', FeatureSelector(columns=mycols))
                     ])

pipeline.fit_transform(df)[:5]