jlengrand.github.io/_posts/2012-02-29-profiling-a-python-script.markdown

layout, status, published, title, author, author_login, author_email, author_url, wordpress_id, wordpress_url, date, categories, tags, comments

layout	status	published	title	author	author_login	author_email	author_url	wordpress_id	wordpress_url	date	categories	tags	comments
post	publish	true	Profiling a Python Script	Julien Lengrand-Lambert	jlengrand	julien@lengrand.fr	http://www.lengrand.fr	546	http://www.lengrand.fr/?p=546	2012-02-29 11:43:08.000000000 +01:00	Python	project euler profile profiler module stack overflow

Here are some simple ways to profile Python scripts.I heavily use this to check my Project Euler solutions.

SOLUTION 1:

The main common option would be to use the profile (or cprofile) module.There are two different ways of using it :

• As a module, by directly running

{% highlight python %} python -m cProfile script.py {% endhighlight %}

• In your code, by importing the utilities

{% highlight python %} import cProfile cProfile.run('function()') # in your __main__ {% endhighlight %}

Bonus : You can use several options for sorting results using the -s switch (cumulative/name/time/file sorting are available).

SOLUTION 2:

If you want to avoid using a command line, or you don't have the profile module installed; here os another possibility. There is also the timeit module available.

{% highlight python %} import timeit t1 = timeit.Timer("function()", "from main import function") print t1.timeit(1) {% endhighlight %}

I use this option on Eclipse because I didn't want to install the profile module on Windows.

This is however less clear, and way less detailed while still useful :).

Choose you profiler option and get on Project Euler !

[Stack Overflow]