Reindexing a series returns NaNs in Pandas

Question

The following code returns a series with NaNs everywhere:

s = pd.Series([1, 2, 3, 4, 5, 6],
                 index=pd.MultiIndex.from_product([["A", "B"], ["c", "d", "e"]]))

s.reindex([('E', 'g'), ('E', 'h'), ('E', 'i'), ('F', 'g'), ('F', 'h'), ('F', 'i')])

or

s.reindex(pd.MultiIndex.from_product([['E', 'F'], ['g', 'h', 'i']]))

How can I reindex the series and retain the original values?

score 3 · Accepted Answer · answered Jun 13 '18 at 13:44

3

This is not reindex , that is change the index

s.index=pd.MultiIndex.from_product([['E', 'F'], ['g', 'h', 'i']])
s
Out[362]: 
E  g    1
   h    2
   i    3
F  g    4
   h    5
   i    6
dtype: int64

answered Jun 13 '18 at 13:44

BENY

317,841
20
164
234

Yeah, "reindex" means "look up these values," not "replace the index." – John Zwinck Jun 13 '18 at 13:53

score 1 · Answer 2 · answered Jun 13 '18 at 13:56

1

If need set new values to second level use MultiIndex.set_levels:

s.index = s.index.set_levels(['g', 'h', 'i'], level=1)
print (s)
A  g    1
   h    2
   i    3
B  g    4
   h    5
   i    6
dtype: int64

answered Jun 13 '18 at 13:56

jezrael

822,522
95
1,334
1,252

Reindexing a series returns NaNs in Pandas

2 Answers2