I have a big DataFrame with 10 columns. I want to sort all rows just for specific columns (two). For example, if this be my data frame
A B C
0 5 1 8
1 8 2 2
2 9 3 3
I want it to sort it just for A and B but for rows so the answer should be like :
A B C
0 1 5 8
1 2 8 2
2 3 9 3
Thank you.
Call np.sort on that specific subslice of columns and assign it back using loc:
# df.loc[:, ['A', 'B']] = np.sort(df.loc[:, ['A', 'B']], axis=1)
df.loc[:, ['A', 'B']] = np.sort(df.loc[:, ['A', 'B']])
df
A B C
0 1 5 8
1 2 8 2
2 3 9 3
I am using sort
s=df[['A','B']]
s.values.sort()
df.update(s)
df
A B C
0 1 5 8
1 2 8 2
2 3 9 3
