I am trying to learn multiprocessing in python but it does not work in its very early steps. for example in the code below:
counter = 2
def train_func():
counter1 = counter*2
p1 = mp.Process(target=train_func)
p2 = mp.Process(target=train_func)
p1.start()
p2.start()
p1.join()
p2.join()
print(counter1)
the result is that NameError: name 'counter1' is not defined. It seems that it does not enter the function. what is wrong with that?
If you have to use some simple functions the easyest way is to use a Pool. Your code will look somethig like:
from multiprocessing import Pool
def f(x):
return x*x
results = []
def saveres(res):
results.append(res)
pool = Pool(nJobs)
print pool.map(f, range(10))
for i in range(10):
pool.apply_async(f, (i,), callback=saveres)
results that is being appended to. multiprocessing does not share state, it is distinct python processes