Sorting rows in the following unique manner (values for columns can be interchanged within the same row, to sort the row)

user_13279071 0
Viewed 48

Input Data frame:

 1.      0th col   1st_col   2nd_col
 2.       23         46         6
 3.       33          56         3
 4.       243        2          21

The output data frame should be like: Index 

 1.    0th col    1st_col     2nd_col
 2.      6          23          46  
 3.      3          33          56
 4.      2          21          243

The rows have to be sorted in ascending or descending order, Independent of columns, Means values for columns can be interchanged within the same row, to sort the row. Sorting rows in the following unique manner. Please Help, I am in the middle of something very important.

pandas-groupby pandas dataframe
1 Answers

Convert DataFrame to numpy array and sort by np.sort with axis=1, then create DataFrame by constructor:

df1 = pd.DataFrame(np.sort(df.to_numpy(), axis=1), 
                   index=df.index, 
                   columns=df.columns)
print (df1)
   0th col  1st_col  2nd_col
1        6       23       46
2        3       33       56
3        2       21      243
user_29010020
Related