为什么从带有map和预定义函数的python文件中调用时，而不是通过带有map和lambda表达式的cli调用时，timeit较慢？

Question

I am checking whether lambda expressions are faster compared to predefined functions. Here is my code to test predefined function

import timeit


def ad(x):
    return x + 2


def test():
    xs = range(10)
    map(ad, xs)

print timeit.timeit("test()", setup="from __main__ import ad, test")

Running it

python add.py
3.21725106239

Changed it to a lambda expression as follows

import timeit


def test():
    xs = range(10)
    map(lambda x: x+2, xs)

print timeit.timeit("test()", setup="from __main__ import test")

The speed is more or less the same, that is ~3 sec+.

However when I run from the command line the following code the results are as follows

➜  /tmp python -mtimeit -s'xs=range(10)' 'map(lambda x: x+2, xs)'
100000 loops, best of 3: 2.42 usec per loop

That is nearly a second lower.

Answer 1

In your commandline version, you're creating the xs list in your setup code, so it's creation is not getting timed. In the non-commandline version, xs is created within test which is the function being timed. Because of this, the two versions aren't timing the same thing.

• Commandline : Timing only map(...)
• File : Timing list(10) and map(...)