CODEWITHSUNDEEP
pythonpandas
Mitch
Mitch

Reputation: 596

How to select rows with NaN in multiple columns without knowing which ones?

Let's say I have a dataframe:

    0   1   2   3   4   5
0   1A  2   3   4   nan 6
1   1   2   3   nan 5   nan
2   1A  2   nan nan nan nan
3   1   2   nan nan nan nan
4   1   2   3   4   5   6

I want to delete rows with a condition in df[0] and where columns 2: are nan. I can't get the 2nd part.

I try to select the rows to drop using:

df.loc[:,2:].isnull()

but that selects rows with any nulls in the last columns, I want the rows where all the columns from 2: are null

I don't want to name the columns because the dataframe won't always have the same number of columns. The only solution I'm thinking of now would be to determine the number of columns in the dataframe, then use

df.loc[x,2].isnull() & df.loc[x,3].isnull() ...

which seems clunky. Ideally I want a way to check if there are nulls in all columns after column 2 (df.loc[x,2:])

Upvotes: 1

Views: 4657

Answers (1)

Mitch
Mitch

Reputation: 596

Quang Hong gave me an answer in a comment that is what I was looking for:

df.loc[:,2:].isnull().all(1)

Upvotes: 3

Related Questions