Sklearn 预处理 - PolynomialFeatures - 如何保留输出数组/数据帧的列名/标题

Question

TLDR：如何从 sklearn.preprocessing.PolynomialFeatures() 函数中获取输出 numpy 数组的标头？

---

假设我有以下代码...

import pandas as pd
import numpy as np
from sklearn import preprocessing as pp

a = np.ones(3)
b = np.ones(3) * 2
c = np.ones(3) * 3

input_df = pd.DataFrame([a,b,c])
input_df = input_df.T
input_df.columns=['a', 'b', 'c']

input_df

    a   b   c
0   1   2   3
1   1   2   3
2   1   2   3

poly = pp.PolynomialFeatures(2)
output_nparray = poly.fit_transform(input_df)
print output_nparray

[[ 1.  1.  2.  3.  1.  2.  3.  4.  6.  9.]
 [ 1.  1.  2.  3.  1.  2.  3.  4.  6.  9.]
 [ 1.  1.  2.  3.  1.  2.  3.  4.  6.  9.]]

如何让 3x10 矩阵/ output_nparray 继承 a、b、c 标签与上述数据的关系？

Answer 1

这行得通：

def PolynomialFeatures_labeled(input_df,power):
    '''Basically this is a cover for the sklearn preprocessing function. 
    The problem with that function is if you give it a labeled dataframe, it ouputs an unlabeled dataframe with potentially
    a whole bunch of unlabeled columns. 

    Inputs:
    input_df = Your labeled pandas dataframe (list of x's not raised to any power) 
    power = what order polynomial you want variables up to. (use the same power as you want entered into pp.PolynomialFeatures(power) directly)

    Ouput:
    Output: This function relies on the powers_ matrix which is one of the preprocessing function's outputs to create logical labels and 
    outputs a labeled pandas dataframe   
    '''
    poly = pp.PolynomialFeatures(power)
    output_nparray = poly.fit_transform(input_df)
    powers_nparray = poly.powers_

    input_feature_names = list(input_df.columns)
    target_feature_names = ["Constant Term"]
    for feature_distillation in powers_nparray[1:]:
        intermediary_label = ""
        final_label = ""
        for i in range(len(input_feature_names)):
            if feature_distillation[i] == 0:
                continue
            else:
                variable = input_feature_names[i]
                power = feature_distillation[i]
                intermediary_label = "%s^%d" % (variable,power)
                if final_label == "":         #If the final label isn't yet specified
                    final_label = intermediary_label
                else:
                    final_label = final_label + " x " + intermediary_label
        target_feature_names.append(final_label)
    output_df = pd.DataFrame(output_nparray, columns = target_feature_names)
    return output_df

output_df = PolynomialFeatures_labeled(input_df,2)
output_df

    Constant Term   a^1 b^1 c^1 a^2 a^1 x b^1   a^1 x c^1   b^2 b^1 x c^1   c^2
0               1   1   2   3   1           2           3   4           6   9
1               1   1   2   3   1           2           3   4           6   9
2               1   1   2   3   1           2           3   4           6   9

Answer 2

工作示例，全部在一行中（我假设“可读性”不是这里的目标）：

target_feature_names = ['x'.join(['{}^{}'.format(pair[0],pair[1]) for pair in tuple if pair[1]!=0]) for tuple in [zip(input_df.columns,p) for p in poly.powers_]]
output_df = pd.DataFrame(output_nparray, columns = target_feature_names)

更新：正如@OmerB 指出的，现在您可以使用get_feature_names method：

>> poly.get_feature_names(input_df.columns)
['1', 'a', 'b', 'c', 'a^2', 'a b', 'a c', 'b^2', 'b c', 'c^2']

Answer 3

scikit-learn 0.18 添加了一个漂亮的get_feature_names() 方法！

>> input_df.columns
Index(['a', 'b', 'c'], dtype='object')

>> poly.fit_transform(input_df)
array([[ 1.,  1.,  2.,  3.,  1.,  2.,  3.,  4.,  6.,  9.],
       [ 1.,  1.,  2.,  3.,  1.,  2.,  3.,  4.,  6.,  9.],
       [ 1.,  1.,  2.,  3.,  1.,  2.,  3.,  4.,  6.,  9.]])

>> poly.get_feature_names(input_df.columns)
['1', 'a', 'b', 'c', 'a^2', 'a b', 'a c', 'b^2', 'b c', 'c^2']

请注意，您必须为其提供列名，因为 sklearn 不会自行从 DataFrame 中读取它。

Answer 4

get_feature_names() 方法很好，但它会将所有变量返回为'x1'、'x2'、'x1 x2'、...等。下面是一个快速将get_feature_names() 输出转换为格式为'Col_1'、'Col_2'、'Col_1 x Col_2' 的列名列表的函数：

在：

def PolynomialFeatureNames(sklearn_feature_name_output, df):
"""
This function takes the output from the .get_feature_names() method on the PolynomialFeatures 
instance and replaces values with df column names to return output such as 'Col_1 x Col_2'

sklearn_feature_name_output: The list object returned when calling .get_feature_names() on the PolynomialFeatures object
df: Pandas dataframe with correct column names
"""
import re
cols = df.columns.tolist()
feat_map = {'x'+str(num):cat for num, cat in enumerate(cols)}
feat_string = ','.join(sklearn_feature_name_output)
for k,v in feat_map.items():
    feat_string = re.sub(fr"\b{k}\b",v,feat_string)
return feat_string.replace(" "," x ").split(',')  

interaction = PolynomialFeatures(degree=2)
X_inter = interaction.fit_transform(input_df)

names = PolynomialFeatureNames(interaction.get_feature_names(),input_df)

print(pd.DataFrame(X_inter, columns= names))

输出：

            1       a       b       c     a^2   a x b   a x c     b^2   b x c  \
0 1.00000 1.00000 2.00000 3.00000 1.00000 2.00000 3.00000 4.00000 6.00000   
1 1.00000 1.00000 2.00000 3.00000 1.00000 2.00000 3.00000 4.00000 6.00000   
2 1.00000 1.00000 2.00000 3.00000 1.00000 2.00000 3.00000 4.00000 6.00000   

      c^2  
0 9.00000  
1 9.00000  
2 9.00000