cpython/Doc/library/timeit.rst

245 lines
8.9 KiB
ReStructuredText
Raw Normal View History

2007-08-15 11:28:22 -03:00
:mod:`timeit` --- Measure execution time of small code snippets
===============================================================
.. module:: timeit
:synopsis: Measure the execution time of small code snippets.
.. index::
single: Benchmarking
single: Performance
This module provides a simple way to time small bits of Python code. It has both
command line as well as callable interfaces. It avoids a number of common traps
for measuring execution times. See also Tim Peters' introduction to the
"Algorithms" chapter in the Python Cookbook, published by O'Reilly.
The module defines the following public class:
.. class:: Timer(stmt='pass', setup='pass', timer=<timer function>)
2007-08-15 11:28:22 -03:00
Class for timing execution speed of small code snippets.
The constructor takes a statement to be timed, an additional statement used for
setup, and a timer function. Both statements default to ``'pass'``; the timer
Merged revisions 73004,73439,73496,73509,73529,73564,73576-73577,73595-73596,73605 via svnmerge from svn+ssh://pythondev@svn.python.org/python/trunk ........ r73004 | jeffrey.yasskin | 2009-05-28 22:44:31 -0500 (Thu, 28 May 2009) | 5 lines Fix nearly all compilation warnings under Apple gcc-4.0. Tested with OPT="-g -Wall -Wstrict-prototypes -Werror" in both --with-pydebug mode and --without. There's still a batch of non-prototype warnings in Xlib.h that I don't know how to fix. ........ r73439 | benjamin.peterson | 2009-06-15 19:29:31 -0500 (Mon, 15 Jun 2009) | 1 line don't mask encoding errors when decoding a string #6289 ........ r73496 | vinay.sajip | 2009-06-21 12:37:27 -0500 (Sun, 21 Jun 2009) | 1 line Issue #6314: logging.basicConfig() performs extra checks on the "level" argument. ........ r73509 | amaury.forgeotdarc | 2009-06-22 14:33:48 -0500 (Mon, 22 Jun 2009) | 2 lines #4490 Fix sample code run by "python -m xml.sax.xmlreader" ........ r73529 | r.david.murray | 2009-06-23 13:02:46 -0500 (Tue, 23 Jun 2009) | 4 lines Fix issue 5230 by having pydoc's safeimport check to see if the import error was thrown from itself in order to decide if the module can't be found. Thanks to Lucas Prado Melo for collaborating on the fix and tests. ........ r73564 | amaury.forgeotdarc | 2009-06-25 17:29:29 -0500 (Thu, 25 Jun 2009) | 6 lines #2016 Fix a crash in function call when the **kwargs dictionary is mutated during the function call setup. This even gives a slight speedup, probably because tuple allocation is faster than PyMem_NEW. ........ r73576 | benjamin.peterson | 2009-06-26 18:37:06 -0500 (Fri, 26 Jun 2009) | 1 line document is_declared_global() ........ r73577 | benjamin.peterson | 2009-06-27 09:16:23 -0500 (Sat, 27 Jun 2009) | 1 line link to extensive generator docs in the reference manual ........ r73595 | ezio.melotti | 2009-06-27 18:45:39 -0500 (Sat, 27 Jun 2009) | 1 line stmt and setup can contain multiple statements, see #5896 ........ r73596 | ezio.melotti | 2009-06-27 19:07:45 -0500 (Sat, 27 Jun 2009) | 1 line Fixed a wrong apostrophe ........ r73605 | georg.brandl | 2009-06-28 07:10:18 -0500 (Sun, 28 Jun 2009) | 1 line Remove stray pychecker directive. ........
2009-06-28 14:22:03 -03:00
function is platform-dependent (see the module doc string). *stmt* and *setup*
may also contain multiple statements separated by ``;`` or newlines, as long as
they don't contain multi-line string literals.
2007-08-15 11:28:22 -03:00
To measure the execution time of the first statement, use the :meth:`timeit`
method. The :meth:`repeat` method is a convenience to call :meth:`timeit`
multiple times and return a list of results.
The *stmt* and *setup* parameters can also take objects that are callable
without arguments. This will embed calls to them in a timer function that
will then be executed by :meth:`timeit`. Note that the timing overhead is a
little larger in this case because of the extra function calls.
2007-08-15 11:28:22 -03:00
.. method:: Timer.print_exc(file=None)
2007-08-15 11:28:22 -03:00
Helper to print a traceback from the timed code.
Typical use::
t = Timer(...) # outside the try/except
try:
t.timeit(...) # or t.repeat(...)
except:
t.print_exc()
The advantage over the standard traceback is that source lines in the compiled
template will be displayed. The optional *file* argument directs where the
traceback is sent; it defaults to ``sys.stderr``.
.. method:: Timer.repeat(repeat=3, number=1000000)
2007-08-15 11:28:22 -03:00
Call :meth:`timeit` a few times.
This is a convenience function that calls the :meth:`timeit` repeatedly,
returning a list of results. The first argument specifies how many times to
call :meth:`timeit`. The second argument specifies the *number* argument for
:func:`timeit`.
.. note::
It's tempting to calculate mean and standard deviation from the result vector
and report these. However, this is not very useful. In a typical case, the
lowest value gives a lower bound for how fast your machine can run the given
code snippet; higher values in the result vector are typically not caused by
variability in Python's speed, but by other processes interfering with your
timing accuracy. So the :func:`min` of the result is probably the only number
you should be interested in. After that, you should look at the entire vector
and apply common sense rather than statistics.
.. method:: Timer.timeit(number=1000000)
2007-08-15 11:28:22 -03:00
Time *number* executions of the main statement. This executes the setup
statement once, and then returns the time it takes to execute the main statement
a number of times, measured in seconds as a float. The argument is the number
of times through the loop, defaulting to one million. The main statement, the
setup statement and the timer function to be used are passed to the constructor.
.. note::
Merged revisions 59259-59274 via svnmerge from svn+ssh://pythondev@svn.python.org/python/trunk ........ r59260 | lars.gustaebel | 2007-12-01 22:02:12 +0100 (Sat, 01 Dec 2007) | 5 lines Issue #1531: Read fileobj from the current offset, do not seek to the start. (will backport to 2.5) ........ r59262 | georg.brandl | 2007-12-01 23:24:47 +0100 (Sat, 01 Dec 2007) | 4 lines Document PyEval_* functions from ceval.c. Credits to Michael Sloan from GHOP. ........ r59263 | georg.brandl | 2007-12-01 23:27:56 +0100 (Sat, 01 Dec 2007) | 2 lines Add a few refcount data entries. ........ r59264 | georg.brandl | 2007-12-01 23:38:48 +0100 (Sat, 01 Dec 2007) | 4 lines Add test suite for cmd module. Written by Michael Schneider for GHOP. ........ r59265 | georg.brandl | 2007-12-01 23:42:46 +0100 (Sat, 01 Dec 2007) | 3 lines Add examples to the ElementTree documentation. Written by h4wk.cz for GHOP. ........ r59266 | georg.brandl | 2007-12-02 00:12:45 +0100 (Sun, 02 Dec 2007) | 3 lines Add "Using Python on Windows" document, by Robert Lehmann. Written for GHOP. ........ r59271 | georg.brandl | 2007-12-02 15:34:34 +0100 (Sun, 02 Dec 2007) | 3 lines Add example to mmap docs. Written for GHOP by Rafal Rawicki. ........ r59272 | georg.brandl | 2007-12-02 15:37:29 +0100 (Sun, 02 Dec 2007) | 2 lines Convert bdb.rst line endings to Unix style. ........ r59274 | georg.brandl | 2007-12-02 15:58:50 +0100 (Sun, 02 Dec 2007) | 4 lines Add more entries to the glossary. Written by Jeff Wheeler for GHOP. ........
2007-12-02 11:22:16 -04:00
By default, :meth:`timeit` temporarily turns off :term:`garbage collection`
during the timing. The advantage of this approach is that it makes
independent timings more comparable. This disadvantage is that GC may be
an important component of the performance of the function being measured.
If so, GC can be re-enabled as the first statement in the *setup* string.
For example::
2007-08-15 11:28:22 -03:00
timeit.Timer('for i in range(10): oct(i)', 'gc.enable()').timeit()
The module also defines two convenience functions:
2007-08-15 11:28:22 -03:00
.. function:: repeat(stmt='pass', setup='pass', timer=<default timer>, repeat=3, number=1000000)
2007-08-15 11:28:22 -03:00
Create a :class:`Timer` instance with the given statement, setup code and timer
function and run its :meth:`repeat` method with the given repeat count and
*number* executions.
.. function:: timeit(stmt='pass', setup='pass', timer=<default timer>, number=1000000)
2007-08-15 11:28:22 -03:00
Create a :class:`Timer` instance with the given statement, setup code and timer
function and run its :meth:`timeit` method with *number* executions.
Command Line Interface
----------------------
When called as a program from the command line, the following form is used::
python -m timeit [-n N] [-r N] [-s S] [-t] [-c] [-h] [statement ...]
Where the following options are understood:
.. program:: timeit
.. cmdoption:: -n N, --number=N
2007-08-15 11:28:22 -03:00
how many times to execute 'statement'
.. cmdoption:: -r N, --repeat=N
2007-08-15 11:28:22 -03:00
how many times to repeat the timer (default 3)
.. cmdoption:: -s S, --setup=S
statement to be executed once initially (default ``pass``)
.. cmdoption:: -t, --time
2007-08-15 11:28:22 -03:00
use :func:`time.time` (default on all platforms but Windows)
.. cmdoption:: -c, --clock
2007-08-15 11:28:22 -03:00
use :func:`time.clock` (default on Windows)
.. cmdoption:: -v, --verbose
2007-08-15 11:28:22 -03:00
print raw timing results; repeat for more digits precision
.. cmdoption:: -h, --help
2007-08-15 11:28:22 -03:00
print a short usage message and exit
A multi-line statement may be given by specifying each line as a separate
statement argument; indented lines are possible by enclosing an argument in
quotes and using leading spaces. Multiple :option:`-s` options are treated
similarly.
If :option:`-n` is not given, a suitable number of loops is calculated by trying
successive powers of 10 until the total time is at least 0.2 seconds.
The default timer function is platform dependent. On Windows,
:func:`time.clock` has microsecond granularity but :func:`time.time`'s
granularity is 1/60th of a second; on Unix, :func:`time.clock` has 1/100th of a
second granularity and :func:`time.time` is much more precise. On either
platform, the default timer functions measure wall clock time, not the CPU time.
This means that other processes running on the same computer may interfere with
the timing. The best thing to do when accurate timing is necessary is to repeat
the timing a few times and use the best time. The :option:`-r` option is good
for this; the default of 3 repetitions is probably enough in most cases. On
Unix, you can use :func:`time.clock` to measure CPU time.
.. note::
There is a certain baseline overhead associated with executing a pass statement.
The code here doesn't try to hide it, but you should be aware of it. The
baseline overhead can be measured by invoking the program without arguments.
The baseline overhead differs between Python versions! Also, to fairly compare
older Python versions to Python 2.3, you may want to use Python's :option:`-O`
option for the older versions to avoid timing ``SET_LINENO`` instructions.
Examples
--------
Here are two example sessions (one using the command line, one using the module
interface) that compare the cost of using :func:`hasattr` vs.
:keyword:`try`/:keyword:`except` to test for missing and present object
attributes. ::
% timeit.py 'try:' ' str.__bool__' 'except AttributeError:' ' pass'
100000 loops, best of 3: 15.7 usec per loop
% timeit.py 'if hasattr(str, "__bool__"): pass'
100000 loops, best of 3: 4.26 usec per loop
% timeit.py 'try:' ' int.__bool__' 'except AttributeError:' ' pass'
1000000 loops, best of 3: 1.43 usec per loop
% timeit.py 'if hasattr(int, "__bool__"): pass'
100000 loops, best of 3: 2.23 usec per loop
::
>>> import timeit
>>> s = """\
... try:
... str.__bool__
... except AttributeError:
... pass
... """
>>> t = timeit.Timer(stmt=s)
>>> print("%.2f usec/pass" % (1000000 * t.timeit(number=100000)/100000))
2007-08-15 11:28:22 -03:00
17.09 usec/pass
>>> s = """\
... if hasattr(str, '__bool__'): pass
... """
>>> t = timeit.Timer(stmt=s)
>>> print("%.2f usec/pass" % (1000000 * t.timeit(number=100000)/100000))
2007-08-15 11:28:22 -03:00
4.85 usec/pass
>>> s = """\
... try:
... int.__bool__
... except AttributeError:
... pass
... """
>>> t = timeit.Timer(stmt=s)
>>> print("%.2f usec/pass" % (1000000 * t.timeit(number=100000)/100000))
2007-08-15 11:28:22 -03:00
1.97 usec/pass
>>> s = """\
... if hasattr(int, '__bool__'): pass
... """
>>> t = timeit.Timer(stmt=s)
>>> print("%.2f usec/pass" % (1000000 * t.timeit(number=100000)/100000))
2007-08-15 11:28:22 -03:00
3.15 usec/pass
To give the :mod:`timeit` module access to functions you define, you can pass a
``setup`` parameter which contains an import statement::
def test():
"Stupid test function"
L = [i for i in range(100)]
2007-08-15 11:28:22 -03:00
if __name__=='__main__':
from timeit import Timer
t = Timer("test()", "from __main__ import test")
print(t.timeit())
2007-08-15 11:28:22 -03:00