How can calculate the time of execution of a command in sage

[faicel]

I look for the command allowing to have the time of execution of a
command in sage



Thank you in advance

[Kevin Horton]

On 30-Jun-09, at 10:58 , faicel wrote:

> I look for the command allowing to have the time of execution of a
> command in sage

sage: timeit?

...
Docstring:

Time execution of a command or block of commands. Displays the
best WALL TIME for execution of the given code. This is
based on
the Python timeit module, which avoids a number of common traps
for measuring execution times. It is also based on IPython's
%timeout command.

TYPICAL INPUT FORMAT:
timeit(statement, preparse=None, number=0, repeat=3,
precision=3)

EXAMPLES:
sage: timeit('2^10000')
625 loops, best of 3: ... per loop

...

--
Kevin Horton
Ottawa, Canada

[William Stein]

On Tue, Jun 30, 2009 at 4:58 PM, faicel<faice...@gmail.com> wrote:
>
> I look for the command  allowing to have the time  of execution of a
> command in sage
>

Also, use the "time" command, e.g.,

sage: time 2+2
CPU times: user 0.00 s, sys: 0.00 s, total: 0.00 s
Wall time: 0.00 s
4
sage:

william

[pang]

Watch also for "cputime" and "walltime", which are very general and
easy to use. Sometimes "time" and "timeit" are not convenient to use
if you have more than one statement.

[William Stein]

One nice trick: If you're in the notebook and you put

%time

as the first line of a notebook cell, then the entire block of code is timed.

William

[Gustavo Rama]

But how con you get the time of execution in a variable?

On Jul 2, 10:33 am, William Stein <wst...@gmail.com> wrote:

[Simon King]

Hi Gustavo!

On 15 Jul., 03:17, Gustavo Rama <gdr...@gmail.com> wrote:
> But how con you get the time of execution in a variable?

Using cputime or walltime: No problem, since they return a number.

Using timeit:

Remember that you could obtain a timing doing
sage: timeit('f=factorial(400).factor()')
625 loops, best of 3: 955 µs per loop

In order to assign the result of the timing to a variable, do

sage: T = timeit.eval('f=factorial(400).factor()')

Then, the necessary information is contained in the attribute 'stats'
of T:

sage: T.stats
(625, 3, 3, 953.62396240234375, '\xc2\xb5s')

So, the first entry gives the number of loops, the second the number
of runs, the third I don't understand, the fourth gives the
computation time, with units given by the last entry. Anyway:

sage: print "%d loops, best of %d: %.*g %s per loop" % T.stats
625 loops, best of 3: 954 µs per loop

Note that this only gives the walltime. Note also that it is not so
nice to compare different timings: It might be that one timing
measures the time in microseconds and the other in milliseconds - IMHO
this is an oddity.

Cheers,
Simon

[Gustavo Rama]

Thanks, I'll try it.
Cheers Gustavo

On Jul 14, 10:42 pm, Simon King <simon.k...@uni-jena.de> wrote:
> Hi Gustavo!
>
> On 15 Jul., 03:17, Gustavo Rama <gdr...@gmail.com> wrote:
>
> > But how con you get the time of execution in a variable?
>
> Using cputime or walltime: No problem, since they return a number.
>
> Using timeit:
>
> Remember that you could obtain a timing doing
>   sage: timeit('f=factorial(400).factor()')

>   625 loops, best of 3: 955 µs per loop

>
> In order to assign the result of the timing to a variable, do
>
>   sage: T = timeit.eval('f=factorial(400).factor()')
>
> Then, the necessary information is contained in the attribute 'stats'
> of T:
>
>   sage: T.stats
>   (625, 3, 3, 953.62396240234375, '\xc2\xb5s')
>
> So, the first entry gives the number of loops, the second the number
> of runs, the third I don't understand,  the fourth gives the
> computation time, with units given by the last entry. Anyway:
>
>   sage: print "%d loops, best of %d: %.*g %s per loop" % T.stats

>   625 loops, best of 3: 954 µs per loop

[Gustavo Rama]

The third entry gives the precicion.

[Robert Bradshaw]

If you want to time running something just once, see cputime() and
walltime().

- Robert