添加两个 pandas dataframe 列，其区别仅在于后缀参数，例如“A_x”、“A_y”，并将这两列重命名为“A”

Question

How to add two pandas dataframe columns which differs by only suffix parameter for eg, "A_x", "A_y" and rename these two columns addition with "A".

For eg, I have a data like this enter image description here

The columns must be renamed without any of the suffix ie., to CT_1 or CT_2 etc....

Answer 1

Use:

df = pd.DataFrame([np.arange(6)], columns=['a','s','CT_1_x','CT_1_y','CT_2_x','CT_2_y'])
print (df)
   a  s  CT_1_x  CT_1_y  CT_2_x  CT_2_y
0  0  1       2       3       4       5

df = df.set_index(['a','s']).groupby(lambda x: x.rsplit('_', 1)[0], axis=1).sum().reset_index()
print (df)
   a  s  CT_1  CT_2
0  0  1     5     9

Answer 2

To add the two columns

df['A'] = df['A_x'] + df['A_y']

and if you want to remove the original columns

df.drop(columns = ['A_x','A_y'])

If you have too many such columns col2sum = ['A_1', 'A_2', ...] to type by hand, the best way would be to melt the df into a long form.

dfm = melt(df, id_vars = ???, value_vars = col2sum)

and go from there (eg groupby ).