官术网_书友最值得收藏!

Evaluating the time taken by a command in IPython

The %timeit magic and the %%timeit cell magic (which applies to an entire code cell) allow us to quickly evaluate the time taken by one or several Python statements. The next recipes in this chapter will show methods for more extensive profiling.

How to do it...

We are going to estimate the time taken to calculate the sum of the inverse squares of all positive integer numbers up to a given n.

  1. Let's define n:
    >>> n = 100000
  2. Let's time this computation in pure Python:
    >>> %timeit sum([1. / i**2 for i in range(1, n)])
    21.6 ms ± 343 μs per loop (mean ± std. dev. of 7 runs,
        10 loops each)
  3. Now, let's use the %%timeit cell magic to time the same computation written on two lines:
    >>> %%timeit s = 0.
        for i in range(1, n):
            s += 1. / i**2
    22 ms ± 522 μs per loop (mean ± std. dev. of 7 runs,
        10 loops each)
  4. Finally, let's time the NumPy version of this computation:
    >>> import numpy as np
    >>> %timeit np.sum(1. / np.arange(1., n) ** 2)
    160 μs ± 959 ns per loop (mean ± std. dev. of 7 runs,
        10000 loops each)

Here, the NumPy vectorized version is 137x faster than the pure Python version.

How it works...

The %timeit command accepts several optional parameters. One such parameter is the number of statement evaluations. By default, this number is chosen automatically so that the %timeit command returns within a few seconds in most cases. However, this number can be specified directly with the -r and -n parameters. Type %timeit? in IPython to get more information.

The %%timeit cell magic also accepts an optional setup statement in the first line (on the same line as %%timeit), which is executed but not timed. All variables created in this statement are available inside the cell.

There's more...

If you are not in an IPython interactive session or in a Jupyter Notebook, you can use import timeit; timeit.timeit(). This function benchmarks a Python statement stored in a string. IPython's %timeit magic command is a convenient wrapper around timeit(), useful in an interactive session. For more information on the timeit module, refer to https://docs.python.org/3/library/timeit.html.

See also

  • The Profiling your code easily with cProfile and IPython recipe
  • The Profiling your code line-by-line with line_profiler recipe
主站蜘蛛池模板: 龙江县| 大连市| 兴文县| 海原县| 荥阳市| 西乌| 清涧县| 汝阳县| 巴青县| 阳城县| 虹口区| 桓仁| 怀来县| 云浮市| 崇左市| 普格县| 北流市| 双峰县| 凯里市| 泸溪县| 杂多县| 霍山县| 伊吾县| 南部县| 绍兴县| 毕节市| 澎湖县| 吉安县| 达拉特旗| 手机| 瑞昌市| 遵义县| 嘉义县| 巍山| 丽水市| 菏泽市| 双流县| 驻马店市| 彝良县| 沐川县| 东明县|