python - How to multiply pandas dataframes and preserve row keys -

i'm struggling multiplying dataframes , preserving row keys.

i have 2 files, call them f1 , f2. f1 has multi-part group key (g1,g2,g3), two-part type key (k1,k2) , weights (r1,r2). f2 has series of values each type key.

i'd join them on k1 , k2, , multiply r1 , r2 each n.

i'm thinking groupby , dataframe multiply should work can't see how it. thing i've got work merge , multiply column column, it's super-slow.

f1 g1  g2  g3  k1  k2  r1  r2           1   2         b   3   4     b     b   2   3  f2 k1  k2  n   r1  r2     1   0   1     2   1   1     3   1   0   b   1   3   4   b   2   4   4   b   3   4   3   c   1   1   1   c   3   4   5   c   2   3   4  result g1  g2  g3  k1  k2  n   r1  r2           1   0   2           2   1   2           3   1   0         b   1   9   16         b   2   12  16         b   3   12  12     b     b   1   6   12     b     b   2   8   12     b     b   3   8   9 

thanks

mrg = f1.merge(f2, on=['k1', 'k2'])  mrg['r1'] = mrg.filter(like='r1').prod(1) mrg['r2'] = mrg.filter(like='r2').prod(1) drops = ['r1_x', 'r1_y', 'r2_x', 'r2_y'] mrg.drop(drops, axis=1)