locate the numeric position of a non numeric index value

Question

Consider the series s below:

s = pd.Series(np.arange(18, 0, -3), list('ABCDEF'))
s

A    18
B    15
C    12
D     9
E     6
F     3
dtype: int32

I want to get the numeric position of 'D'

This will do it, but I think we can all agree this is gross:

s.reset_index().index.to_series()[s.reset_index().iloc[:, 0] == 'D'].iloc[0]

score 8 · Accepted Answer · answered Jul 18 '16 at 16:45

8

print(s.index.get_loc('D'))
3

answered Jul 18 '16 at 16:45

jezrael

1

Looks like the built-in for such a task, lovely! – Divakar Jul 18 '16 at 16:48
2

Note: this returns an array if there is more than one 'D' in the index! – Alex Jul 18 '16 at 16:48

Divakar · Answer 2 · 2016-07-18T16:43:19.517

4

np.searchsorted(s.index.values,'D')

Or use the method, like so -

s.index.searchsorted('D')

edited Jul 18 '16 at 16:43

answered Jul 18 '16 at 16:37

Divakar

I love that you provided the numpy solution then found the pandas solution afterwards ;-). You should add the numpy tag to my post. – piRSquared Jul 18 '16 at 16:44
1

@piRSquared That's the beauty of answering questions, specially with tags that I am really not familiar with! ;) – Divakar Jul 18 '16 at 16:45

score 3 · Answer 3 · answered Jul 18 '16 at 16:45

3

m = s.index == 'D'
idx = m.argmax() if m.any() else None

answered Jul 18 '16 at 16:45

Alex

3 Answers3