Python Multiprocessing: Example not loading on Jupyter or IPython

Question

I am following Berkeley's Python Numerical Methods book and I am trying to run some code for the multiprocessor. The setup is straightforward (showing the time difference between serial and parallel version), but I can't get the parallel version to run. I saw one answer that said that a different library might be needed to run multithreading in Jupyter Notebooks. I have found a few errors in this book so I wouldn't be surprised that this could need some updating. I tried this in an IPython environment and also in Jupyter Lab and I didn't get either to load (AttributeError: Can't get attribute 'random_square' on <module 'main' (built-in)>).

import multiprocessing as mp
print(f"Number of cpu: {mp.cpu_count()}")


# Serial version - works fine

import numpy as np
import time

def random_square(seed):
    np.random.seed(seed)
    random_num = np.random.randint(0, 10)
    return random_num**2

t0 = time.time()
results = []
for i in range(10000000): 
    results.append(random_square(i))
t1 = time.time()
print(f'Execution time {t1 - t0} s')



# Parallel version - doesn't compute

import multiprocessing as mp

t0 = time.time()
n_cpu = mp.cpu_count()

pool = mp.Pool(processes=n_cpu)
results = [pool.map(random_square, range(10000000))]
t1 = time.time()
print(f'Execution time {t1 - t0} s')

Answer 1

I found the answer. You can reference this previous question to see various solutions. The one I went with was installing a new package multiprocess and I used map_async instead of the map function and it worked like a charm!