Closed
Description
Code Sample
import pandas as pd
data1={(1, 2): 3,
(4, None): 6}
s1 = pd.Series(data=data1)
data2={(1, 2): 3,
(None, 5): 6}
s2 = pd.Series(data=data2)
Problem description
s1
has MultiIndex while s2
has Index of tuple just because the second element of the data2
has None
as the first element of the key tuple.
>>> s1
1 2 3.0
4 NaN NaN
dtype: float64
>>> s2
(1, 2) 3
(None, 5) 6
dtype: int64
Is this a bug? or is this how it's supposed to be?
FYI, both of mi1
and mi2
below yield MultiIndex as I expect.
mi1 = pd.MultiIndex.from_tuples([(1, 2), (4, None)])
mi2 = pd.MultiIndex.from_tuples([(1, 2), (None, 5)])