Changing NaN values to 0 based on the value on other column in pandas DataFrame

Question

I want to change NaN values from one column based on the value of another column.

First table

Targeted table

So I want to change NaN to 0, if the value of column A in that row is also 0.

The python code that I create:

for i in range(len(df)):
    if df.loc[i,'A'] == 0 and math.isnan(df.loc[i,'B']) = True:
        df.loc[i,'B'] = 0

But this code seems so slow, because I want to do this for 20 million rows. Is there any faster way to do this?

user_13138364 · Answer

First combine the conditions using & (element-wise and operator):

conditions = df['B'].isnull() & df['A'].eq(0)

# 0    False
# 1     True
# 2    False
# 3    False
# dtype: bool

Then you can vectorize the assignment (no loop):

Either use a single loc:
```
df.loc[conditions, 'B'] = 0
```
Or mask the values that satisfy those conditions with 0:
```
df['B'] = df['B'].mask(conditions, 0)
```

Output:

#    A    B
# 0  2  3.0
# 1  0  0.0
# 2  2  NaN
# 3  4  2.0