On Sun, Jul 26, 2009 at 12:31 AM, Carlo Hamalainen <
carlo.hamalai...@gmail.com> wrote:
> On Sat, Jul 25, 2009 at 5:08 PM, Stephen Hartke wrote:
> > Might this be related to how binomial is evaluated using GiNaC?
>
> Valgrind says yes:
>
> ==26568== 4 bytes in 1 blocks are definitely lost in loss
On Sat, Jul 25, 2009 at 5:08 PM, Stephen Hartke wrote:
> Might this be related to how binomial is evaluated using GiNaC? Similar
> problems occur when replacing binomial with log.
Valgrind says yes:
==26568== 4 bytes in 1 blocks are definitely lost in loss record 35 of 3,312
==26568==at 0x
On Sat, Jul 25, 2009 at 6:04 PM, Justin C. Walker wrote:
> > I noticed that in the notebook, the code does create a problem,
> ^ not??
> > but random values do.
>
Yes, I missed a "not".
> Yup. I now see what you see: memory usage increases after
On Jul 25, 2009, at 15:08 , Stephen Hartke wrote:
> On Sat, Jul 25, 2009 at 1:54 PM, Justin C. Walker
> wrote:
>
>> I just tried this on 4.0.2 and 4.1 (on Mac OS X, 10.5.7), and got the
>> same values before and after the loop, so something else must be
>> involved.
>>
>
> Justin,
>
> Thanks
On Sat, Jul 25, 2009 at 1:54 PM, Justin C. Walker wrote:
> I just tried this on 4.0.2 and 4.1 (on Mac OS X, 10.5.7), and got the
> same values before and after the loop, so something else must be
> involved.
>
Justin,
Thanks for your response! Did you run it from the command line or the
notebo
On Jul 25, 2009, at 08:08 , Stephen Hartke wrote:
> The following code ends up using a lot of memory:
>
> print get_memory_usage()
> for i in range(10):
>b=binomial(5,2)
> print get_memory_usage()
>
> Output:
> 133.48828125
> 135.015625
I just tried this on 4.0.2 and 4.1 (on Mac OS X, 1
