How can I save result from groupby in a new column?-CodePudding

I have a dataframe

key1	key2	key3	value1	value2
1	a	s2	3	4
1	a	s2	2	3
2	b	j6	1	1

and I want as result

key1	key2	key3	value1	value2	sum_value1	sum_value2
1	a	s2	3	4	5	7
1	a	s2	2	3	5	7
2	b	j6	1	1	1	1

sum_value1 is the summation of values in value1 by grouping key1, key2, key3. And so for sum_value2.

How can I get this? Thank you!

What I used so far:

df["sum_value1"] = df["value1"].groupby(["key1","key2","key3"]).transform('sum')

CodePudding user response：

use groupby and transform to return the sum of individual columns

df[['sum_value1','sum_value2']]=df.groupby(['key1','key2','key3'])[['value1','value2']].transform(sum)
df

    key1    key2     key3   value1  value2  sum_value1  sum_value2
0      1       a       s2      3       4       5          5
1      1       a       s2      2       1       5          5
2      2       b       j6      1       1       1          1