pandas df[df["A"]==None] is not the same as df["A"].values==None

Question

Assume I have a dataframe df where the column A consists of 10 None and the rest is something else.

If I do the slicing df=df[df["A"]==None] I get a wrong result. I figured out that df["A"]==None returns False (even when the elements are None) but df["A"].values==None returns the correct.

How come? Shouldn't we be able to slice in the first way ?

https://stackoverflow.com/questions/45117272/pandas-filtering-none-values — Seb, Nov 27 '19 at 13:03

score 1 · Answer 1 · answered Nov 27 '19 at 13:08

1

You should use isna() method over the serie.

For your case:

df = df.loc[df['A'].isna()]

answered Nov 27 '19 at 13:08

Ignacio Franco

126
1
8

To explain the two different answers, the `df.isna()` and `df.isnull()` do the same thing -- they are aliases – moo Nov 27 '19 at 13:11

score 1 · Answer 2 · answered Nov 27 '19 at 13:08

1

You can use as follows

df = df[df['A'].isnull()]

answered Nov 27 '19 at 13:08

Omrum Cetin

1,320
13
17

Thanks but the question is why the first approach does not work – CutePoison Nov 28 '19 at 20:45

pandas df[df["A"]==None] is not the same as df["A"].values==None

2 Answers2