How to make my python script run on my GPU

Question

I created a script to find a number inside pi:

from math import pi
from mpmath import mp
from time import sleep as sleep


def loop(find):

    #Breaks the find string into a list

    findList = []


    print('Finding ' + str(find))

    num = 1000

    while True:

        mp.dps = num

        string = str(mp.pi)

        result = string.find(str(find))

        if result == -1:

            print("Couldn't find " + str(find) + " within the first " + str(num) + " of Pi. Looking moving into the first " + str(num * 10) + " digits instead")

            num = num * 10

            continue

            pass

        else:

            print(str(find )+ ' was found at character: ' + str(result))
            break

        pass

    pass

def main():

    find = input("What do you want to find: ")


    find = int(find)
    character = loop(find)

    
    


if True:

    main()

    input = ()

When a long number is inputed it takes a long time to process for obvious reasons. I'm running an Intel i5-9300h and a GTX 1650. I am wondering if 1) I can make this code run on my GPU instead of my CPU 2) If so, how do I do this? 3) Would it even benefit performance?

Any help is much appreciated.

Shan S · Accepted Answer · 2020-06-27 10:33:33Z

1

I think numba will help with what you're looking for. It can run python code with CUDA support (i.e. your graphics card). Using CUDA for math is often faster than using a CPU because of better multithreading, however, whether you would get any benefit from it will depend on implementation. In your example I think you'll have to write a function that can approximate pi in a multithreaded way, in order to take advantage of what CUDA has to offer.

answered Jun 27, 2020 at 10:33

Shan S

6985 silver badges18 bronze badges

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

Collectives™ on Stack Overflow

How to make my python script run on my GPU

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related