I have a pandas dataframe like follows,

+------------+-------------------+---------+--------+----------+
| process_no | process_durations | columns | orders | customer |
+------------+-------------------+---------+--------+----------+
| 0          | 3                 | [0]     | [3109] | [0]      |
+------------+-------------------+---------+--------+----------+
| 1          | 100               | [11]    | [5855] | [0]      |
+------------+-------------------+---------+--------+----------+
| 2          | 81                | [8]     | [5304] | [0]      |
+------------+-------------------+---------+--------+----------+

I want to combine this dataframe with itself and get the total length for orders, the length of unique values for columns, and the list of unique values for customers. Then, I want to sort in descending for the orders length and ascending for the columns length. But I should also keep the original version of the combination. For this situation, I did the following,

from itertools import combinations
from operator import itemgetter

vals = data.values.tolist()
cross_combine = list(combinations(vals, r=2))

sorted_cross_combine = sorted(
    [
        (
            x,
            -(len(x[0][3]) + len(x[1][3])),
            len(set(x[0][2] + x[1][2])),
            list(set(x[0][4] + x[1][4]))
        )
        for x in cross_combine
    ],
    key=itemgetter(1, 2)
)
print(sorted_cross_combine)

[(([0, 3, [0], [3109], [0]], [1, 100, [11], [5855], [0]]), -2, 2, [0]),
 (([0, 3, [0], [3109], [0]], [2, 81, [8], [5304], [0]]), -2, 2, [0]),
 (([1, 100, [11], [5855], [0]], [2, 81, [8], [5304], [0]]), -2, 2, [0])]

What I want to ask is,

Can I do this while the combine process is still taking place, provided that it is more effective? For example, I know there is no such function, but I imagine something like this,

def calc(x, y):
    return (
        x + y,
        -(len(x[3]) + len(y[3])),
        len(set(x[2] + y[2])),
        list(set(x[4] + y[4]))
    )

cross_combine = list(combinations(vals, r=2, func=calc))

Or is there a way I can make the whole process more effective? The process of creating the sorted_cross_combine value takes about 20 seconds for a vals list with approximately 6500 elements.

Thanks in advance.