[FIXED] Replacing the first occurrence of a value in each row of a pandas DataFrame

Issue

I have a DataFrame like this:

col1	col2	col3	col4
5	7	12	9
0	9	9	1
9	9	1	1
10	5	2	9
9	3	0	18

Each row has at least one 9 and for each row, I want to replace the first instance of it with 90.

Currently, I'm doing:

out = df.mask(df.eq(9) & df.apply(lambda x: ~x.duplicated(), axis=1), 90)

Is there any better/faster way than this?

Expected output:

   col1  col2  col3  col4
0     5     7    12    90
1     0    90     9     1
2    90     9     1     1
3    10     5     2    90
4    90     3     0    18

Constructor:

data = {'col1': [5, 0, 9, 10, 9],
        'col2': [7, 9, 9, 5, 3],
        'col3': [12, 9, 1, 2, 0],
        'col4': [9, 1, 1, 9, 18]}
df = pd.DataFrame(data)

Solution

You may check with numpy assign

df.values[df.index, np.argmax(df.values==9,1)] = 90 
df
Out[56]: 
   col1  col2  col3  col4
0     5     7    12    90
1     0    90     9     1
2    90     9     1     1
3    10     5     2    90
4    90     3     0    18

Answered By - BENY