when I try:
set(np.array([1,2,3,3,3]))
I get:
{1, 2, 3}
which makes sense to me.
when I try:
set(np.array([1,2,3,3,np.nan, np.nan, np.nan]))
I get
{nan, 1.0, 2.0, 3.0, nan, nan}
any ideas why?
Asked
Active
Viewed 29 times
1
Ezer K
- 3,637
- 3
- 18
- 34
-
3Well... their IDs aren't the same. – cs95 Jan 14 '18 at 09:49
-
On second thought, it seems like the implementation may have changed, `np,nan` at one point would yield the same object but that's changed. – cs95 Jan 14 '18 at 09:51
-
1https://stackoverflow.com/questions/26245862/reducing-pandas-series-with-multiple-nan-values-to-a-set-gives-multiple-nan-valu – Rajat Mishra Jan 14 '18 at 09:52
-
@RajatMishra better dupe, thanks – cs95 Jan 14 '18 at 09:52
-
Thanks all, missed the dupe – Ezer K Jan 14 '18 at 10:51