0

I have 5 lists that contain some data:

a = [52, 265, 1, 98, 26]
b = [42, 85, 45, 2, 3, 8, 632]
c = [7, 731, 92, 65, 28, 62]
d = [3, 5, 44, 55]
e = [15, 35, 850, 6, 18, 41]

What I want to do is get the sum of the values of every possible combination, for example:

a[0] + b[0] + c[0] + d[0] + e[0] = **121**
...
a[4] + b[6] + c[5] + d[3] + e[5] = **816**

This needs to be done is the fastest way possible, because in my code the length of these lists is bigger, thus making the calculation of all possibilities extremely time demanding.

My first idea was using nested for loops, but that's not fast enough. I imagine some transformations and some magic could be done with NumPy to speed up this process.

Improve this question
6
  • Python is the wrong language if you want this to be done in the fastest possible way. Commented Dec 27, 2017 at 18:29
  • 4
    I strongly suspect that you'll need to avoid actually computing all these sums to solve your underlying problem efficiently. Commented Dec 27, 2017 at 18:30
  • 3
    this is a trick question. .. you should not be fully exploring the search space Commented Dec 27, 2017 at 18:32
  • The core idea of the analysis I'm making is to go through all of the possibilities in order to find the very best one, so this is probably the only way of doing that. Commented Dec 27, 2017 at 18:35
  • 1
    This is almost certainly an XY Problem. What do you need all the sums for? What is the end goal? Commented Dec 27, 2017 at 18:49

2 Answers 2

Reset to default
1

the fastest way to actually sum all the combinations(you actually want the product, not the combinations) is to generate them and sum them 

print([(x,sum(x)) for x in itertools.product(my_list1,my_list2,my_list3,...)])

but I suspect that the secret is to not check all the products, but instead to intelligently select specific values to narrow your search space

Improve this answer
Sign up to request clarification or add additional context in comments.

1 Comment

Based on my tests this seems to be the fastest way of doing it, and it's also very simple. The only thing I changed was (x,sum(x)) to sum(x). Thanks, Joran.
0

You want something like this ?

a = [52, 265, 1, 98, 26]
b = [42, 85, 45, 2, 3, 8, 632]
c = [7, 731, 92, 65, 28, 62]
d = [3, 5, 44, 55]
e = [15, 35, 850, 6, 18, 41]

import itertools

print(list(map(lambda x:sum(x),itertools.product(a,b,c,d,e))))

output:

119, 139, 954, 110, 122, 145, 121, 141, 956, 112, 124, 147, 160, 180, 995, 151, 163, 186, 171, 191, 1006, 162, 174, 197, 843, 863, 1678, 834, 846, 869, 845, 865, 1680, 836, 848, 871, 884, 904, 1719, 875, 887, 910, 895, 915, 1730, 886, 898, 921, 204, 224, 1039, 195, 207, 230, 206, 226, 1041, 197, 209, 232, 245, 265, 1080, 236, 248, 271, 256, 276, 1091, 247, 259, 282, 177, 197, 1012, 168, 180, 203, 179, 199, 1014, 170, 182, 205, 218, 238, 1053, 209, 221, 244, 229, ....
Improve this answer

2 Comments

Yes, that's the result I'm looking for, I just don't know if this is the fastest way to achieve it
@CaioCamatta Not fastest way but you can try it.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.