Hi I want to replace blanks in my dataframe as 0. When I run this:

list(df['col'])

I get list of values['12345','78910',...., '' ,'5678', '12344', '', '7654']. How can I replace these '' missing values with 0. I want the whole df to be replaced by 0 when there is a blank like this ''.

CodePudding user response：

As you have strings, I imagine that the best is to convert to_numeric, then fillna with 0 the invalid/missing values:

df['col'] = pd.to_numeric(df['col'], errors='coerce').fillna(0, downcast='infer')

For the whole DataFrame:

df = df.apply(pd.to_numeric, errors='coerce').fillna(0, downcast='infer')

Example output:

     col
0  12345
1  78910
2      0
3   5678
4  12344
5      0
6   7654

Used input:

df = pd.DataFrame({'col': ['12345','78910','' ,'5678', '12344', '', '7654']})

CodePudding user response：

You can try this:

>>>
data = {'a': ['12345', '78910', '', '5678', '12344', '', '7654'],
        'b': ['295', '7', '', '8', '', '37484', '901']}
df = pd.DataFrame(data)
print(df)

    a       b
0   12345   295
1   78910   7
2       
3   5678    8
4   12344   
5           37484
6   7654    901

>>>
out = df.replace('', 0)
print(out)

    a       b
0   12345   295
1   78910   7
2   0       0
3   5678    8
4   12344   0
5   0       37484
6   7654    901