Pandas: `a < b` behaves differently than `a.lt(b)`

Asked Jun 07 '19 at 14:46

Active Jun 07 '19 at 14:47

Viewed 31 times

I have the following series

x = pd.Series([1, 2], index=['a', 'b'])
y = pd.Series([1, 2], index=['b', 'a'])

When I try

x < y

I get an error

ValueError: Can only compare identically-labeled Series objects

However, with

x.lt(y)

it produces the corrent result.

The documentation states that they should be equivalent (with the exception of missing values)

Is this a bug in pandas, or am I misunderstanding something?

asked Jun 07 '19 at 14:46

blue_note

See https://stackoverflow.com/q/25435229/2336654 – piRSquared Jun 07 '19 at 14:48
@piRSquared: must be a previous version, in my case `x > y` crashes – blue_note Jun 07 '19 at 14:50
yikes! what version? – piRSquared Jun 07 '19 at 14:51
@piRSquared: `0.24.2`. at least, failing is better than what happens in your case... – blue_note Jun 07 '19 at 14:52
1

Oh, you mean it raises the ValueError? Yeah, that was the improvement they made. I asked that question almost 5 years ago. If you really didn't want to use `pd.Series.lt` you could align the series first `x, y = x.align(y); x < y` – piRSquared Jun 07 '19 at 14:54

0 Answers0