[Rivet] Quantify the "agreemnet" with Rivet

[Andy Buckley]

On 12/03/14 18:17, Liron Barak wrote:
> Hi Holger, Andy,
> 
> Many thanks! GofLegend=1 worked perfectly :)
> For what Andy suggested, what is Rivet 2? Is that implemented in athena too?

It's what it sounds like: version 2 of Rivet! ATLAS is still using
version 1 in all production releases, but we have almost finished the
Athena migration work. You can use it in the Athena development
nightlies at the moment.

All future versions of Rivet will be in the version 2 series, but as it
uses FastJet3 we unfortunately can't backport it to older (17.*) ATLAS
releases. Once the ATLAS 19.1.0 release is out people will be able to
use Rivet2 on the grid via ATLAS software, and at that point I would
advise only using those releases for running Rivet in ATLAS: the whole
system is much improved, there are more analyses, and it's where the
development and bug-fixing is happening.

Andy


> On Wed, Mar 12, 2014 at 5:09 PM, Andy Buckley <andy.buckley at cern.ch
> <mailto:andy.buckley at cern.ch>> wrote:
> 
>     With Rivet 2 you also have full programmatic access to the histograms
>     via YODA, so it should not be a lot of work to make your own numerical
>     comparisons. The tools for doing this should improve as YODA and the
>     plotting tools develop and become better integrated. Contact us if you
>     need any help with using the YODA Python or C++ interfaces.
> 
>     But the make-plots GoF measure suggested by Holger is definitely the
>     easiest "built-in" way at the moment, although it doesn't give you
>     _programmatic_ access to the quality of fit number.
> 
>     Andy
> 
> 
>     On 12/03/14 15:33, Holger Schulz wrote:
>     > On 12/03/14 14:15, Liron Barak wrote:
>     >> Dear Rivet experts,
>     >>
>     >> I do some MC studies and I use Rivet all the time to compare the
>     >> different MC options with the data, thanks for the nice tool!
>     >> I mainly look on the ratio (to the data) and see the agreement... I
>     >> wonder if Rivet can also give me back some kind of number? A way to
>     >> quantify what is the closest to data not by eye.....
>     >>
>     >> Many thanks
>     >> Liron
>     >>
>     >>
>     >> _______________________________________________
>     >> Rivet mailing list
>     >> Rivet at projects.hepforge.org <mailto:Rivet at projects.hepforge.org>
>     >> http://www.hepforge.org/lists/listinfo/rivet
>     > Hi Liron,
>     >
>     > you could use the built-in feature of make-plots to calculate a
>     GoF measure
>     > and display it in the Legend.
>     >
>     > In order to make this appear, add
>     >
>     > GofLegend=1
>     >
>     > in the PLOT section of the input files for make-plots.
>     >
>     > On the command line you could do this e.g. by running this command:
>     >
>     > for i in *dat;do sed -i "s/Legend=1/Legend=1\nGofLegend=1/g" $i;done
>     >
>     >
>     > Hope that helps,
>     > Holger
>     >
>     >
>     > _______________________________________________
>     > Rivet mailing list
>     > Rivet at projects.hepforge.org <mailto:Rivet at projects.hepforge.org>
>     > http://www.hepforge.org/lists/listinfo/rivet
>     >
> 
> 
>     --
>     Dr Andy Buckley, Royal Society University Research Fellow
>     Particle Physics Expt Group, University of Glasgow / PH Dept, CERN
> 
> 


-- 
Dr Andy Buckley, Royal Society University Research Fellow
Particle Physics Expt Group, University of Glasgow / PH Dept, CERN