将系列索引设置为具有更多级别的另一个索引

Question

我正在尝试使用包含比系列原始索引更多的级别的索引(ix)设置Series(a1)的索引。

>>> a = pd.DataFrame({'a': [1, 2, 3], 'b': ['b', 'b', 'b'], 'x': [4, 5, 6]}).set_index(['a', 'b'])
>>> a
x
a b   
1 b  4
2 b  5
3 b  6
>>> 
>>> a1 = a['x']
>>> a1
a  b
1  b    4
2  b    5
3  b    6
Name: x, dtype: int64
>>> ix = pd.MultiIndex.from_product(([1, 2, 3], ['b', 'c'], [10, 20]), names=['a', 'b', 'c'])
>>> ix
MultiIndex(levels=[[1, 2, 3], [u'b', u'c'], [10, 20]],
labels=[[0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2], [0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1], [0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1]],
names=[u'a', u'b', u'c'])
>>> a.set_index(ix)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "python2.7/site-packages/pandas/core/frame.py", line 3164, in set_index
frame.index = index
File "python2.7/site-packages/pandas/core/generic.py", line 3627, in __setattr__
return object.__setattr__(self, name, value)
File "pandas/_libs/properties.pyx", line 69, in pandas._libs.properties.AxisProperty.__set__
File "python2.7/site-packages/pandas/core/generic.py", line 559, in _set_axis
self._data.set_axis(axis, labels)
File "python2.7/site-packages/pandas/core/internals.py", line 3074, in set_axis
(old_len, new_len))
ValueError: Length mismatch: Expected axis has 3 elements, new values have 12 elements

因此，我期待以下系列：

a  b   c
1  b  10  4
2  b  10  5
3  b  10  6
1  c  10  nan   # [1, c] wasn't an index in a1
2  c  10  nan   # ...
3  c  10  nan   # ...
1  b  20  4     # [1, b] was an index of a1, so use that value
2  b  20  5     # ...
3  b  20  6     # ...
1  c  20  nan   # [1, c] wasn't an index in a1
2  c  20  nan   # ...
3  c  20  nan   # ...
# if there was an index in a1 that isn't in `ix`, it should be maintained with
# its value and the index should be augmented

我怎样才能用熊猫做到这一点？

Answer 1

使用Series.reindex，但输出有点不同，因为MultiIndex排序，如果以后有效工作，显然是必要的 - 来源：

要有效地对MultiIndex对象进行索引和切片，需要对它们进行排序。与任何索引一样，您可以使用 sort_index()。

a = a.reindex(ix)
print (a)
x
a b c      
1 b 10  4.0
20  4.0
c 10  NaN
20  NaN
2 b 10  5.0
20  5.0
c 10  NaN
20  NaN
3 b 10  6.0
20  6.0
c 10  NaN
20  NaN

Answer 2

问题是你有一个空的数据框，它有三列，你试图给它分配十二列的多索引;如果最初创建一个包含四列的空数据框，则错误将消失：

df = pd.DataFrame(pd.np.empty((0, 12)))    
df.columns = pd.MultiIndex(levels = [['first', 'second'], ['a', 'b']], labels = [[0, 0, 1, 1], [0, 1, 0, 1]])

或者，您可以使用多索引创建空数据框，如下所示：

multi_index = pd.MultiIndex(levels = [['first', 'second'], ['a', 'b']], labels = [[0, 0, 1, 1], [0, 1, 0, 1]])    
df = pd.DataFrame(columns=multi_index)
df
#   first    second
#  a    b   a     b