Make things parallel.

pyrallel.py

import math
from concurrent.futures import ProcessPoolExecutor, as_completed
from typing import Callable


def cut_list(data: list, n: int):
    for i in range(1, math.ceil(len(data) / n) + 1):
        yield data[(i - 1) * n: i * n]

def parallelize(func: Callable, inputs: list, n: int):
    results = []
    
    while len(inputs) != 1:
        with ProcessPoolExecutor() as executor:
            processes = [executor.submit(func, chunk) for chunk in cut_list(inputs, n)]

            for process in as_completed(processes):
                results.append(process.result())
            
        inputs = results.copy()
        results = []

    return func(inputs)
    
if __name__ == "__main__":
    parallelize(sum, list(range(1_000_000)), 100_000)