Filling null values in pandas based on value in another column conditionally

Question

how to fill null values in a row based on the value in another column.

A  B
0  5 
1  NAN
1  6
0  NAN

for the null value in B if coressponding value in A is 0 then fill with the previous value.

A  B
0  5 
1  NAN
1  6
0  6

```

it want it to be like this

Possible duplicate of [Pandas fill missing values in dataframe from another dataframe](https://stackoverflow.com/questions/29357379/pandas-fill-missing-values-in-dataframe-from-another-dataframe) — Michael Heidelberg, Apr 18 '19 at 16:58
@MichaelHeidelberg that question asks about fill null values based on a common index, not based on a condition on another column and the previous value. — user3483203, Apr 18 '19 at 17:00

score 3 · Answer 1 · answered Apr 18 '19 at 16:57

3

numpy.where + isnull + ffill

df.assign(
    B=np.where(df.A.eq(0) & df.B.isnull(), df.B.ffill(), df.B)
)

answered Apr 18 '19 at 16:57

user3483203

score 1 · Answer 2 · answered Apr 18 '19 at 17:03

1

Another way using loc,

df.loc[df['A'].eq(0), 'B'] = df['B'].ffill()

    A   B
0   0   5
1   1   NaN
2   1   6
3   0   6

answered Apr 18 '19 at 17:03

Vaishali

sentence · Answer 3 · 2019-04-18T17:28:59.350

1

A faster way (compared to the previous ones):

import pandas as pd
import numpy as np

df = pd.DataFrame({'A':[0,1,1,0], 'B':[5,np.nan,6,np.nan]})

df.B = np.where(df.A==0, df.B.ffill(), df.B)

and you get:

    A   B
0   0   5.0
1   1   NaN
2   1   6.0
3   0   6.0

edited Apr 18 '19 at 17:28

answered Apr 18 '19 at 17:22

sentence

3 Answers3