[Rivet] New CMS plugin: strangeness in UE (CMS-PAS-QCD-11-010) and UE at forward eta (CMS-PAS-FWD-11-003)

[Hannes Jung]

Dear Hendrik

thanks a lot for checking so carefully new analyses which are submitted to Rivet.

However, I do not understand your argumentation: for me and many others
the important point is to have analyses available in Rivet, but I do not care
whether they have 100 or 200 or only 10 lines of code. This might be
important from an programmers point of view, but for me it is just irrelevant.
Of course, the code should run and produce correct results.

If the code is changed by anyone other than the author, who is then
responsible for validation ?

BTW, for non-expert it is not always easy to find out which tools
exist already ...

Cheers

Hannes



On 12.12.2012, at 18:00, Hendrik Hoeth wrote:

Hi Albert,

I've now checked the two analyses, and while the strangeness analysis
was straight forward, I spent several hours and lots of cursing cleaning
up the forward UE analysis. In the future I will simply send such code
back, because we do expect some minimum quality control from you.

You find my code in the attachment -- it is about half as long as the
original code and, unlike yours, doesn't crash after ~1300 Pythia events.

Some "highlights" of the original code:


- A self-written selection sort algorithm to sort particle by rapidity,
 instead using the provided tools. Takes up 17 lines of code and scales
 as O(n^2).


- This could be a "ParticleVector myTempParticles = sfsv.particles();".
 Note also the cast from Particle to Particle.
 ------------------------------------
 ParticleVector myTempParticles;
 foreach (const Particle& p, sfsv.particles()) {
   myTempParticles.push_back(Particle(p));
 }
 ------------------------------------


- I *love* the following one. The whole purpose of the loop is to find
 the hardest jet (i.e. jets[0]). And the "if (index_1 != -1)" is
 naturally fulfilled, because we require at least one jet:
 ------------------------------------
 const FastJets& jetpro = applyProjection<FastJets>(event, "Jets");
 const Jets& jets = jetpro.jetsByPt(1.0*GeV);

 if (jets.size()>0) {

   int index_1 = -1;        // for the jet with the 1.highest pT
   double tempmax_1   = -100;

   // jet with the 1.highest pt
   for (size_t ijets = 0; ijets<jets.size(); ++ijets) {
     if (tempmax_1 == -100 || tempmax_1 < jets[ijets].momentum().pT()/GeV) {
       tempmax_1 = jets[ijets].momentum().pT()/GeV;
       index_1   = ijets;
     }
   }

   if (index_1 != -1) {
     // do something, fill a histogram
   }
 }
 ------------------------------------
 So this code boils down to:
 ------------------------------------
 const FastJets& jetpro = applyProjection<FastJets>(event, "Jets");
 const Jets& jets = jetpro.jetsByPt(1.0*GeV);
 if (jets.size()<1) vetoEvent;
 // do something, fill a histogram, using jets[0].
 ------------------------------------


- In your code all the analysis code had three identical copies: One
 for each energy.


- I wonder what invisible particles the jet finder is supposed to use,
 after being given only charged particles:
 ------------------------------------
 const ChargedFinalState fsj(-2.5,2.5,0.3*GeV);
 addProjection(fsj, "FSJ");

 FastJets jetpro (fsj, FastJets::ANTIKT, 0.5);
 jetpro.useInvisibles();
 addProjection(jetpro, "Jets");
 ------------------------------------


- Casting a FourMomentum to a FourMomentum doesn't do much, except for
 rendering the code unreadable. And why the author does a manual
 mass-square just one line below using the mass2() method is beyond my
 mental capabilities.
 ------------------------------------
 FourMomentum Xfourmom;

 for (int ipart=0; ipart <= deltaymax_pos; ++ipart) {
   Xfourmom += myRapiditySortedParticles[ipart].momentum();
 }
 if(FourMomentum(Xfourmom).mass2() <0 ) vetoEvent;

 long double Mx2 = FourMomentum(Xfourmom).mass()*FourMomentum(Xfourmom).mass();
 ------------------------------------


This list goes on and on. Global variables that could be local. Strange
types ("signed int", "long double", ...). PLEASE have a look at the code
next time before dumping it on us, because we will have to maintain it
in the future -- together with several hundred other analyses. Thanks.

Cheers,

   Hendrik

--
If your dreams don't scare you, then you are not dreaming big enough.
<CMS_2012_PAS_FWD_11_003.cc>_______________________________________________
Rivet mailing list
Rivet at projects.hepforge.org<mailto:Rivet at projects.hepforge.org>
http://www.hepforge.org/lists/listinfo/rivet

***********************************************************************
Hannes Jung
Email: Hannes.Jung at desy.de<mailto:Hannes.Jung at desy.de>
mobile :+49 40 8998 93741
http://www.desy.de/~jung
Tel: +49 (0) 40 8998 3741
Fax: +49 (0) 40 8998 3093
DESY, CMS 01B/02.213
Notkestr.85, 22603 Hamburg, FRG
***********************************************************************




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.hepforge.org/lists-archive/rivet/attachments/20121213/0302824f/attachment.html>