VBF filters for QCD sample requested by VBF InvHiggs group

[chayanit]

They have been merged to CMSSW_8_X_X and now I just backported them to CMSSW_7_1_X. Thanks

[cmsbuild]

A new Pull Request was created by @chayanit (Chayanit Asawatangtrakuldee) for CMSSW_7_1_X.

It involves the following packages:

GeneratorInterface/GenFilters

@vciulli, @covarell, @thuer, @cmsbuild, @bendavid, @davidlange6, @govoni can you please review it and eventually sign? Thanks.
@mkirsano this is something you requested to watch as well.
@slava77, @Degano, @smuzaffar you are the release manager for this.

Following commands in first line of a comment are recognized

• +1|approve[d]|sign[ed]: L1/L2's to approve it
• -1|reject[ed]: L1/L2's to reject it
• assign <category>[,<category>[,...]]: L1/L2's to request signatures from other categories
• unassign <category>[,<category>[,...]]: L1/L2's to remove signatures from other categories
• hold: L1/all L2's/release manager to mark it as on hold
• unhold: L1/user who put this PR on hold
• merge: L1/release managers to merge this request
• [@cmsbuild,] please test: L1/L2 and selected users to start jenkins tests
• [@cmsbuild,] please test with cms-sw/cmsdist#<PR>: L1/L2 and selected users to start jenkins tests using externals from cmsdist

[govoni]

+1

[cmsbuild]

The tests are being triggered in jenkins.
https://cmssdt.cern.ch/jenkins/job/ib-any-integration/10118/console

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next CMSSW_7_1_X IBs after it passes the integration tests. This pull request requires discussion in the ORP meeting before it's merged. @slava77, @davidlange6, @Degano, @smuzaffar

[cmsbuild]

+1
Tested at: aebccdd
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-12657/10118/summary.html

[cmsbuild]

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-12657/10118/summary.html

[davidlange6]

@chayanit - why is this include needed?

[chayanit]

There was a usage of "sqrt" before but not anymore. This include can be removed. @davidlange6

[chayanit]

Hi @davidlange6 could you move this request forward? I think they are all fine right? thanks.

[srimanob]

@davidlange6
Which part of the code do we still have issue? I think we expect to have large QCD production using this filter?
cc @jimbrooke: Here is what filter we expect to use.

[davidlange6]

@chayanit @srimanob - sorry for the delay, I think the ~minimal change that would make me happier is below. Much better would be to simply remove filterGenJets in favor of adding a few lines to the double loop over jets.

[dlange@lxplus0096 GenFilters]$ git diff
diff --git a/GeneratorInterface/GenFilters/interface/VBFGenJetFilter.h b/GeneratorInterface/GenFilters/interface/VBFGenJetFilter.h
index a0dbe69..0b88d5d 100644
--- a/GeneratorInterface/GenFilters/interface/VBFGenJetFilter.h
+++ b/GeneratorInterface/GenFilters/interface/VBFGenJetFilter.h
@@ -41,7 +41,7 @@ private:
void printGenVector(std::vectorHepMC::GenParticle* vec);
double nuMET(std::vectorHepMC::GenParticle* vNu);

• std::vector<const reco::GenJet*> filterGenJets(const std::vectorreco::GenJet* jets);
• std::vector<const reco::GenJet*> filterGenJets(const std::vectorreco::GenJet &jets);
  // std::vector<const reco::GenJet*> filterGenJets(const std::vectorreco::GenJet* jets);

//**************************
diff --git a/GeneratorInterface/GenFilters/src/VBFGenJetFilter.cc b/GeneratorInterface/GenFilters/src/VBFGenJetFilter.cc
index 3ff6c21..1a492a8 100644
--- a/GeneratorInterface/GenFilters/src/VBFGenJetFilter.cc
+++ b/GeneratorInterface/GenFilters/src/VBFGenJetFilter.cc
@@ -4,9 +4,6 @@

#include <HepMC/GenVertex.h>

-// ROOT includes

-#include "TMath.h"

// C++ includes
#include

@@ -35,13 +32,13 @@ VBFGenJetFilter::~VBFGenJetFilter(){

}

-vector<const reco::GenJet*> VBFGenJetFilter::filterGenJets(const vectorreco::GenJet* jets){
+vector<const reco::GenJet*> VBFGenJetFilter::filterGenJets(const vectorreco::GenJet &jets){

vector<const reco::GenJet*> out;

• for(unsigned i=0; isize(); i++){
• for(unsigned i=0; i<jets.size(); i++){
• const reco::GenJet* j = &((*jets)[i]);
• const reco::GenJet* j = &(jets[i]);

if(j->p4().pt() >ptMin && j->p4().eta()>etaMin && j->p4().eta()<etaMax)
{
@@ -60,7 +57,7 @@ bool VBFGenJetFilter::filter(edm::Event& iEvent, const edm::EventSetup& iSetup)

Handle< vectorreco::GenJet > handleGenJets;
iEvent.getByToken(m_inputTag_GenJetCollection, handleGenJets);

• const vectorreco::GenJet* genJets = handleGenJets.product();
• const vectorreco::GenJet &genJets = *(handleGenJets.product());

// Getting filtered generator jets
vector<const reco::GenJet*> filGenJets = filterGenJets(genJets);
@@ -74,17 +71,10 @@ bool VBFGenJetFilter::filter(edm::Event& iEvent, const edm::EventSetup& iSetup)
const reco::GenJet* pA = filGenJets[a];
const reco::GenJet* pB = filGenJets[b];

•  // Getting the dijet vector
  

- math::XYZTLorentzVector diJet = pA->p4() + pB->p4();

   // Testing opposite hemispheres
   double dijetProd = pA->p4().eta()*pB->p4().eta();
   if(oppositeHemisphere && dijetProd>=0){continue;}

•  // Testing dijet mass
  

•  double invMass = diJet.mass();
  

- if(invMass<=minInvMass || invMass>maxInvMass){continue;}

• // Testing dijet delta eta
  double dEta = fabs(pA->p4().eta()-pB->p4().eta());
  if(dEta<=minDeltaEta || dEta>maxDeltaEta){continue;}
  @@ -92,6 +82,14 @@ bool VBFGenJetFilter::filter(edm::Event& iEvent, const edm::EventSetup& iSetup)
  // Testing dijet delta phi
  double dPhi = fabs(reco::deltaPhi(pA->p4().phi(),pB->p4().phi()));
  if(dPhi<=minDeltaPhi || dPhi>maxDeltaPhi){continue;}
  +

•  // Getting the dijet vector
  

•  math::XYZTLorentzVector diJet = pA->p4() + pB->p4();
  

•  // Testing dijet mass
  

•  double invMass = diJet.mass();
  

•  if(invMass<=minInvMass || invMass>maxInvMass){continue;}
  

• return true;
  }

[chayanit]

Hi @davidlange6,

Thanks for your reply. Let me link you to @joaopela who is the owner of these codes. He has some questions about your comments.

Best,
Chayanit

[joaopela]

I agree with the removing of the TMath.h and the inclusion of " a<filGenJets.size()-1". But i do not understand the rest of the requests.

1. Why, would I want to remove the cut on the dijet invariant mass? This is exactly one of the cuts that we want to require on sample generation/filtering.

2. The presence of a separated filterGenJets optimises the number of operations per event. Lets look at the example with 20 genJets with only 3 that pass pT and eta requirements:

=> with filterGenJets filtering operations:
20x compare pTmin, etaMinm etaMax (with filterGenJets)
3x mjj calc and dijet var compare (over selected jets only, since combinations of 2 over 3 is 3 possible pairs)

=> without filterGenJets filtering operations (Conbinations of 2 over 20 is 190 possible pairs):
190x (2x one per jet) compare pTmin, etaMinm etaMax
190x mjj calc and dijet var compare

Clearly, pre filtering the generator jets reduces event-per-event computation time.

[joaopela]

I forgot to tag @davidlange6 in the last post.

[davidlange6]

@chayanit @joaopela - sorry for the delay - I'll clarify so that this can be iterated on in the development release while this PR is approved since I have not had time to follow up.

1. Why, would I want to remove the cut on the dijet invariant mass? This is exactly one of the cuts that >we want to require on sample generation/filtering.

its not - just moved lower as its the expensive calc to make

2. The presence of a separated filterGenJets optimises the number of operations per event. Lets look at >the example with 20 genJets with only 3 that pass pT and eta requirements:

Right,but function calls are also expensive. anyway a detail.