I have the following dataset df
UniqueID Col1 Col2
0 1234 5 NaN
1 1235 3 4
2 1233 NaN 3
3 1111 3 NaN
I'd like to know the number of rows where Col1 is not null and Col2 is null.
Add a comment
|
4 Answers
(df.Col1.notnull() & df.Col2.isnull()).sum()
2
Comments
I'd obviously go with PiRSquared's.
If you, however, want to go something fancy with query, then use
In [430]: df.query('Col1 == Col1 & Col2 != Col2').shape[0]
Out[430]: 2
Comments
By using dropna
In [451]: df.dropna(axis=0,how='any',subset=['Col1']).Col2.isnull().sum()
Out[451]: 2
Comments
First off, in your example 'nAn' is not a null. So, let's replace that string with np.nan.
df = df.replace('nAn',pd.np.nan)
df.Col1.notnull().sum()
or
df.Col1.count() # Note: `count` ignores nulls where `size`, `shape`, and `len` do not.
3
And, use isnull to check for nulls explicityly:
df.Col2.isnull().sum()
2
2 Comments
Zero
I think, OP needs on both columns together, see PiRSquared's.
Scott Boston
Ah.. will modify later