Pandas:系列长度不相等的布尔值索引 [英] Pandas: boolean indexing with unequal Series lengths
问题描述
给出两个熊猫系列对象A和Matchs.匹配项包含索引A的子集,并且具有布尔项.等效于逻辑索引的是什么?
Given two pandas series objects A and Matches. Matches contains a subset of the indexes of A and has boolean entries. How does one do the equivalent of logical indexing?
如果Matches与A的长度相同,则可以使用:
If Matches were the same length as A, one could just use:
A[Matches] = 5.*Matches
匹配项少于A的人得到:
With Matches shorter than A one gets:
error: Unalignable boolean Series key provided
按要求绘制插图
In [15]: A = pd.Series(range(10))
In [16]: A
Out[16]: 0 0
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
dtype: int64
In [17]: Matches = (A<3)[:5]
In [18]: Matches
Out[18]: 0 True
1 True
2 True
3 False
4 False
dtype: bool
In [19]: A[Matches] = None
---------------------------------------------------------------------------
IndexingError Traceback (most recent call last)
<ipython-input-19-7a04f32ce860> in <module>()
----> 1 A[Matches] = None
C:\Anaconda\lib\site-packages\pandas\core\series.py in __setitem__(self, key, value)
631
632 if _is_bool_indexer(key):
--> 633 key = _check_bool_indexer(self.index, key)
634 try:
635 self.where(~key, value, inplace=True)
C:\Anaconda\lib\site-packages\pandas\core\indexing.py in _check_bool_indexer(ax, key)
1379 mask = com.isnull(result.values)
1380 if mask.any():
-> 1381 raise IndexingError('Unalignable boolean Series key provided')
1382
1383 result = result.astype(bool).values
IndexingError: Unalignable boolean Series key provided
In [20]:
我正在寻找的结果是:
In [16]: A
Out[16]: 0 None
1 None
2 None
3 3
4 4
5 5
6 6
7 7
8 8
9 9
dtype: int64
Matches系列的结构是人造的,仅供参考.另外,在我的情况下,行索引显然是非数字的,并且不等于元素值...
The construction of the Matches series is artificial and for illustration only. Also, in my case row indexes are obviously non-numeric and not equal to element values...
推荐答案
好吧,您没有想要的东西,因为对于没有任何内容的系列,int64
可能不是dtype.没有一个不是整数.但是您可以接近:
Well, you can't have what you want, because int64
is not a possible dtype for a series containing None. None isn't an integer. But you can get close:
>>> A = pd.Series(range(10))
>>> Matches = (A<3)[:5]
>>> A[Matches[Matches].index] = None
>>> A
0 None
1 None
2 None
3 3
4 4
5 5
6 6
7 7
8 8
9 9
dtype: object
之所以起作用,是因为Matches[Matches]
选择了Matches
正确的元素.
Which works because Matches[Matches]
selects the elements of Matches
which are true.
这篇关于Pandas:系列长度不相等的布尔值索引的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!