Thank you very much Thomas. This is the kind of explanation I needed!
Janek
Dnia czwartek, 18 października 2012, Thomas Schilling napisał:
> On 18 October 2012 13:15, Janek S. wrote:
> >> Something like this might work, not sure what the canonical way is.
> >> (...)
> >
> > This is basically the
On 18 October 2012 13:15, Janek S. wrote:
>> Something like this might work, not sure what the canonical way is.
>> (...)
>
> This is basically the same as the answer I was given on SO. My concerns about
> this solutions are:
> - rnf requires its parameter to belong to NFData type class. This is
Yes, Criterion always discards the time of the first evaluation.
On 18 October 2012 15:06, Janek S. wrote:
>> So the evaluation will be included in the benchmark, but if "bench" is
>> doing enough trials it will be statistical noise.
> When I intentionally delayed my dataBuild function (using del
On Thu, Oct 18, 2012 at 9:06 AM, Janek S. wrote:
>> So the evaluation will be included in the benchmark, but if "bench" is
>> doing enough trials it will be statistical noise.
>
> When I intentionally delayed my dataBuild function (using delayThread
> 100) the estimated time
> of benchmark wa
> So the evaluation will be included in the benchmark, but if "bench" is
> doing enough trials it will be statistical noise.
When I intentionally delayed my dataBuild function (using delayThread 100)
the estimated time
of benchmark was incorrect, but when I got the final results all runs were
On Thu, Oct 18, 2012 at 4:23 AM, Janek S. wrote:
> Dear list,
>
> during past few days I spent a lot of time trying to figure out how to write
> Criterion benchmarks,
> so that results don't get skewed by lazy evaluation. I want to benchmark
> different versions of an
> algorithm doing numerical
> Something like this might work, not sure what the canonical way is.
> (...)
This is basically the same as the answer I was given on SO. My concerns about
this solutions are:
- rnf requires its parameter to belong to NFData type class. This is not the
case for some data
structures like Repa ar
I don't know if you have already read them,
but Tibell's slides on High Performance Haskell are pretty good:
http://www.slideshare.net/tibbe/highperformance-haskell
There is a section at the end where he runs several tests using Criterion.
HTH,
A.
On 18 October 2012 11:45, Claude Heiland-Allen
Hi Janek,
On 18/10/12 10:23, Janek S. wrote:
during past few days I spent a lot of time trying to figure out how to write
Criterion benchmarks,
so that results don't get skewed by lazy evaluation. I want to benchmark
different versions of an
algorithm doing numerical computations on a vector.
Dear list,
during past few days I spent a lot of time trying to figure out how to write
Criterion benchmarks,
so that results don't get skewed by lazy evaluation. I want to benchmark
different versions of an
algorithm doing numerical computations on a vector. For that I need to create
an inpu
10 matches
Mail list logo
