Comparativo entre set Intersection e buscas em diferentes representações de dados Python3.

teste_intersection.py

#!/usr/bin/python3

import random
import time

row_count = 10000000

my_list = {random.randint(0, 2000): random.randint(0, 255) for i in range(row_count)}
my_list2 = {random.randint(0, 2000): random.randint(0, 255) for i in range(row_count)}

def teste():
    start = time.time()
    result = {}
    
    # dict
    for x in my_list:
        if x in my_list2:
            result[x] = my_list[x]

    end = time.time()
    t1 = end - start
    
    # set com intersection
    start = time.time()
    result = {}
    my_list_set = set(my_list)
    my_list2_set = set(my_list2)
    
    inter = my_list_set.intersection(my_list2_set)
    for x in inter:
        result[x] = my_list[x]

    end = time.time()
    t2 = end - start

    # set direto
    start = time.time()
    result = {}
    my_list_set = set(my_list)
    my_list2_set = set(my_list2)
    for x in my_list_set:
        if x in my_list2_set:
            result[x] = my_list[x]

    end = time.time()
    t3 = end - start

    # tupla
    start = time.time()
    result = {}
    my_list_set = set(my_list)
    my_list2_tuple = tuple(my_list2)

    for x in my_list_set:
        if x in my_list2_tuple:
            result[x] = my_list[x]

    end = time.time()
    t4 = end - start

    # lista    
    start = time.time()
    result = {}
    my_list_set = set(my_list)
    my_list2_list = list(my_list2)

    for x in my_list_set:
        if x in my_list2_list:
            result[x] = my_list[x]

    end = time.time()
    t5 = end - start

    return t1, t2, t3, t4, t5

t1 = 0
t2 = 0
t3 = 0
t4 = 0
t5 = 0

n_tests = 1000

for x in range(0, n_tests):
    aux1, aux2, aux3, aux4, aux5 = teste()
    t1 += aux1
    t2 += aux2
    t3 += aux3
    t4 += aux4
    t5 += aux5

print("Media de tempo For normal DICT: %s " % (t1/n_tests))
print("Media de tempo SET: %s " % (t2/n_tests))
print("Media de tempo SET IF: %s " % (t3/n_tests))
print("Media de tempo TUPLE: %s " % (t4/n_tests))
print("Media de tempo LIST: %s " % (t5/n_tests))