python 字典查询比列表快_字典检索和列表索引哪个更快？

如果不需要搜索，请使用列表。它比dict更快，占用的内存更少。对于小型收集(100个项目)，速度差异最小，但对于大型收集，dict将慢20%左右。它肯定会使用更多的内存。在

下面是一些比较list和dict访问速度的timeit代码。它还显示了收集对象本身(但不是它们所保存的数据对象的RAM)消耗的RAM。在#!/usr/bin/env python3

''' Compare list vs dict access speed

See http://stackoverflow.com/q/39192442/4014959

Written by PM 2Ring 2016.08.29

'''

from sys import getsizeof

from timeit import Timer

commands = {'dict' : 'for i in r:d[i]', 'list' : 'for i in r:a[i]'}

def time_test(loops, reps):

timings = []

setup = 'from __main__ import r, a, d'

for name, cmd in commands.items():

result = Timer(cmd, setup).repeat(reps, loops)

result.sort()

timings.append((result, name))

timings.sort()

for result, name in timings:

print(name, result)

#Find the ratio of slowest / fastest

tlo, thi = [timings[i][0][0] for i in (0, -1)]

print('ratio: {0:f}\n'.format(thi / tlo))

num = 2000

r = range(num)

a = list(r)

d = {i:i for i in r}

fmt = 'Sizes: num={}, list={}, dict={}'

print(fmt.format(num, getsizeof(a), getsizeof(d)))

loops, reps = 2000, 3

time_test(loops, reps)

输出

^{pr2}$

速度差实际上比这些结果中显示的要高，因为从r范围对象检索整数所用的时间与执行list&dict访问所用的时间大致相同。您可以通过在commands字典中添加这样一个条目来度量：'none': 'for i in r:i'