python中如何产生交互项？

sklearn里的PolynomialFeatures可以基本实现这个功能

from sklearn.preprocessing import PolynomialFeatures
# 新建一个dataframe
X = pd.DataFrame()
X['x1'] = [1, 1, 1, 2, 0, 0]
X['x2'] = [3, 2, 1, 3, 2, 1]

下面构造交互项

interaction = PolynomialFeatures(2, interaction_only=True, include_bias=False)
X_transformed = interaction.fit_transform(X)

得到的X_transformed就有三列，第三列是x1和x2的乘积

array([[ 1.,  3.,  3.],
       [ 1.,  2.,  2.],
       [ 1.,  1.,  1.],
       [ 2.,  3.,  6.],
       [ 0.,  2.,  0.],
       [ 0.,  1.,  0.]])

直接手算似乎更方便

df['x1_x2'] = df['x1'] * df['x2']