如何为已有多级列的DataFrame添加新的列层级？

原始DataFrame结构 #

x   
   A  B
0  0  1
1  2  3
2  4  5
3  6  7
4  8  9

尝试的代码 #

x.columns = pd.MultiIndex.from_product([['D'], x.columns])

错误信息 #

Traceback (most recent call last):
  File "C:\Users\adel.moustafa\DashBoard\main.py", line 262, in <module>
    calculate_yield()
  File "C:\Users\adel.moustafa\DashBoard\main.py", line 204, in calculate_yield
    Analyzer.yield_analyzer_by(yield_data, all_data_df, df_info['P/F Criteria'], 'batch')
  File "C:\Users\adel.moustafa\DashBoard\Modules\Analyzer.py", line 163, in yield_analyzer_by
    x.columns = pd.MultiIndex.from_product([['D'], x.columns])
  File "C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\indexes\multi.py", line 621, in from_product
    codes, levels = factorize_from_iterables(iterables)
  File "C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\arrays\categorical.py", line 2881, in factorize_from_iterables
    codes, categories = zip(*(factorize_from_iterable(it) for it in iterables))
  File "C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\arrays\categorical.py", line 2881, in <genexpr>
    codes, categories = zip(*(factorize_from_iterable(it) for it in iterables))
  File "C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\arrays\categorical.py", line 2854, in factorize_from_iterable
    cat = Categorical(values, ordered=False)
  File "C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\arrays\categorical.py", line 451, in __init__
    dtype = CategoricalDtype(categories, dtype.ordered)
  File "C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\dtypes\dtypes.py", line 183, in __init__
    self._finalize(categories, ordered, fastpath=False)
  File "C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\dtypes\dtypes.py", line 337, in _finalize
    categories = self.validate_categories(categories, fastpath=fastpath)
  File "C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\dtypes\dtypes.py", line 530, in validate_categories
    if categories.hasnans:
  File "pandas\_libs\properties.pyx", line 37, in pandas._libs.properties.CachedProperty.__get__
  File "C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\indexes\base.py", line 2681, in hasnans
    return bool(self._isnan.any())
  File "pandas\_libs\properties.pyx", line 37, in pandas._libs.properties.CachedProperty.__get__
  File "C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\indexes\base.py", line 2666, in _isnan
    return isna(self)
  File "C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\dtypes\missing.py", line 144, in isna
    return _isna(obj)
  File "C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\dtypes\missing.py", line 169, in _isna
    raise NotImplementedError("isna is not defined for MultiIndex")
NotImplementedError: isna is not defined for MultiIndex

可复现代码 #

import pandas as pd
import numpy as np
x = pd.DataFrame(np.arange(10).reshape(5, 2), columns=pd.MultiIndex.from_product([['x'], ['A', 'B']]))
x.columns = pd.MultiIndex.from_product([['D'], x.columns])

pd.MultiIndex.from_product要求传入的是单层可迭代序列，但你传入的x.columns本身是MultiIndex多层索引对象。内部处理时会尝试对这个MultiIndex执行isna检查，而pandas不支持直接对MultiIndex调用isna方法，因此抛出错误。

给已有的多级列添加顶层，推荐两种方法：

方法一：使用insert方法直接插入新层级 #

# 在第0层（顶层）插入新元素'D'
x.columns = x.columns.insert(0, 'D', level=0)

方法二：构造新的元组列表生成MultiIndex #

将原有列的每个元组前拼接新层级元素，再转为MultiIndex：

new_columns = [('D',) + col for col in x.columns]
x.columns = pd.MultiIndex.from_tuples(new_columns)

执行后，新的列结构为：

D   
   x   
   A  B
0  0  1
1  2  3
2  4  5
3  6  7
4  8  9

内容的提问来源于stack exchange，提问作者Adel Moustafa