In python, is there a simple way to pass 1+ args to pool.map

Question

I want use pool.map to simplify concurrency in my scripts and I find the same question as below link, but not get a concise method for this issue.

Python multiprocessing pool.map for multiple arguments

from multiprocessing.dummy import Pool as ThreadPool

def run_multiple_args(tuple_args):
    args_1 = tuple_args[0]
    args_2 = tuple_args[1]
    print args_1, args_2

pool = ThreadPool(2)

arg_list = [(1,2), (4,8)]

pool.map(func=run_multiple_args, iterable=arg_list)
pool.close()
pool.join()

i want use def run_multiple_args(args_1, args_2): to replace tuple_args.

no 3rd party lib, less code more better. any idea?

You're not the only person who's wanted this: Starting in Python 3.3, there's now a `starmap` method on `Pool` objects that does exactly what you want (`pool.starmap(func, iterable_of_args_tuples)` calls `func(*args)`). — Blckknght, Oct 10 '15 at 05:03

score 0 · Answer 1 · answered Oct 10 '15 at 02:57

0

You can use a lambda to unpack the argument tuple:

from multiprocessing.dummy import Pool as ThreadPool

def run_multiple_args(args_1, args_2):
    print args_1, args_2

pool = ThreadPool(2)
arg_list = [(1,2), (4,8)]

pool.map(func=lambda args: run_multiple_args(*args), iterable=arg_list) # <-- see here
pool.close()
pool.join()

answered Oct 10 '15 at 02:57

kirbyfan64sos

10,377
6
54
75

@dakangz If you liked my answer, you can accept my answer. :) – kirbyfan64sos Oct 10 '15 at 14:52

In python, is there a simple way to pass 1+ args to pool.map

1 Answers1