Selecting rows in pandas dataframe

Question

I would need to select rows satisfying the following conditions:

if (X is True and Z is false) | ( X is false and Z is true) then assign to a new column True as value.

I tried with this:

df[(df[X']==True & df['Z']==False) | (df['X']==False & df['Z']==True)]

but I got the following error:

ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

I have tried using any() as follows

Does this answer your question? [Truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all()](https://stackoverflow.com/questions/36921951/truth-value-of-a-series-is-ambiguous-use-a-empty-a-bool-a-item-a-any-o) — AMC, Sep 04 '20 at 00:32
@AMC while I agree `()`s around the clauses would work, many of these answers also address the fact the this complex set of conditions is unnecessary and go beyond just point out the precedence issues. It is not normal to comment and downvote all the answers for a dupe question. Flag or vote the question. — AChampion, Sep 04 '20 at 01:00
@AChampion _while I agree ()s around the clauses would work, many of these answers also address the fact the this complex set of conditions is unnecessary and go beyond just point out the precedence issues._ Even if we ignore the duplicate issue, the question is still missing a MCVE, and part of the error output. — AMC, Sep 04 '20 at 01:05
_It is not normal to comment and downvote all the answers for a dupe question._ What does _normal_ mean, and according to who? I shared a few posts in [the discussion with BEN_YO](https://chat.stackoverflow.com/rooms/220948/discussion-between-ben-yo-and-amc) which show that it isn't all *that* rare. — AMC, Sep 04 '20 at 01:05

score 0 · Answer 1 · answered Sep 04 '20 at 00:06

0

df['X']==True & df['Z']==False must be (df['X']==True) & (df['Z']==False) (and everything else, respectively). In Python, operator & has a higher precedence than ==. Your expression is interpreted as df['X']==(True & df['Z'])==False.

answered Sep 04 '20 at 00:06

DYZ

55,249
10
64
93

https://stackoverflow.com/q/36921951 – AMC Sep 04 '20 at 00:32

Andy L. · Accepted Answer · 2020-09-04T00:08:36.623

0

If all your values are boolean values as in you conditions, you may try this

df[(df['X'] & ~df['Z']) | (~df['X'] & df['Z'])]

edited Sep 04 '20 at 00:08

answered Sep 04 '20 at 00:06

Andy L.

24,909
4
17
29

1

Doesn't this simplify to `df['X'] != df['Z']`? – AChampion Sep 04 '20 at 00:12
@AChampion: yep, it is correct :) – Andy L. Sep 04 '20 at 00:16
https://stackoverflow.com/q/36921951 – AMC Sep 04 '20 at 00:32

BENY · Answer 3 · 2020-09-04T00:15:02.817

0

Check with

df[((df['X']==True) & (df['Z']==False)) | ((df['X']==False) & (df['Z']==True))]

Also a little math here

df[df.X.astype(int).add(df.Z.astype(int)==1]

From Achampion

df[df.X!=df.Z]

edited Sep 04 '20 at 00:15

answered Sep 04 '20 at 00:07

BENY

317,841
20
164
234

1

Of `df.X != df.Z`? – AChampion Sep 04 '20 at 00:13
@AChampion correct ~ – BENY Sep 04 '20 at 00:14
1

Also I tested and `xor` works too `df.X ^ df.Z` – AChampion Sep 04 '20 at 00:17
https://stackoverflow.com/q/36921951 – AMC Sep 04 '20 at 00:32
Let us [continue this discussion in chat](https://chat.stackoverflow.com/rooms/220948/discussion-between-ben-yo-and-amc). – BENY Sep 04 '20 at 00:41

Sam S. · Answer 4 · 2020-09-04T00:28:05.743

0

If X and Y contain only True and False, the below will give the answer (please check How do you get the logical xor of two variables in Python?):

df[df['X']!=df['Z']]

edited Sep 04 '20 at 00:28

answered Sep 04 '20 at 00:20

Sam S.

627
1
7
23

1

https://stackoverflow.com/q/36921951 – AMC Sep 04 '20 at 00:32

Selecting rows in pandas dataframe

4 Answers4