Run a multiprocessing job with Python 2.7 / Windows

问题

Based on this answer, I'd like to run this multiprocessing job with Python 2.7 / Windows:

def main():
    import itertools as it
    from multiprocessing import Pool

    def dothejob(i, j, k):
        print i, j, k

    the_args = it.product(range(100), range(100), range(100))
    pool = Pool(4)

    def jobWrapper(args): 
        return dothejob(*args)

    res = pool.map(jobWrapper, the_args)

if __name__ == '__main__':
    main()    

The main() and the last two lines are necessary because without them, there's the well known bug:

This probably means that you are on Windows and you have forgotten to use the proper idiom in the main module:

if __name__ == '__main__':
    ....

But even with that, I get this error:

File "C:\Users\User\Desktop\test.py", line 14, in main res = pool.map(jobWrapper, the_args) File "C:\Python27\lib\multiprocessing\pool.py", line 251, in map return self.map_async(func, iterable, chunksize).get() File "C:\Python27\lib\multiprocessing\pool.py", line 558, in get raise self._value cPickle.PicklingError: Can't pickle : attribute lookup >builtin.function failed

Where is there a cPickle involved here and why this error / how to solve it?

回答1:

All definitions must be on module scope:

import itertools as it
from multiprocessing import Pool, freeze_support

def dothejob(i, j, k):
    print i, j, k

def jobWrapper(args): 
    return dothejob(*args)

def main():
    the_args = it.product(range(100), range(100), range(100))
    pool = Pool(4)
    res = pool.map(jobWrapper, the_args)

if __name__ == '__main__':
    freeze_support() #you need this in windows
    main()

You also need freeze_support call in windows

来源：https://stackoverflow.com/questions/48581455/run-a-multiprocessing-job-with-python-2-7-windows