Skip to content

Commit

Permalink
Add second SV collection
Browse files Browse the repository at this point in the history
  • Loading branch information
Vilius Cepaitis committed Nov 12, 2020
1 parent f08f050 commit 337515d
Show file tree
Hide file tree
Showing 6 changed files with 174 additions and 47 deletions.
1 change: 1 addition & 0 deletions DataFormats/interface/ChargedCandidateFeatures.h
View file
Original file line number Diff line number Diff line change
Expand Up @@ -38,6 +38,7 @@ struct ChargedCandidateFeatures
int cpf_matchedMuon;
int cpf_matchedElectron;
int cpf_matchedSV;
int cpf_matchedSV_adapted;
int cpf_track_ndof;

float cpf_dZmin;
Expand Down
1 change: 1 addition & 0 deletions DataFormats/interface/XTagFeatures.h
View file
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,7 @@ class XTagFeatures {
JetFeatures jet_features;
ShallowTagInfoFeatures tag_info_features;
std::vector<SecondaryVertexFeatures> sv_features;
std::vector<SecondaryVertexFeatures> sv_adapted_features;
std::vector<NeutralCandidateFeatures> npf_features;
std::vector<ChargedCandidateFeatures> cpf_features;
std::vector<MuonCandidateFeatures> mu_features;
Expand Down
84 changes: 74 additions & 10 deletions NANOProducer/plugins/NANOProducer.cc
View file
Original file line number Diff line number Diff line change
Expand Up @@ -48,6 +48,7 @@ NANOProducer::NANOProducer(const edm::ParameterSet& iConfig) :
produces<nanoaod::FlatTable>("cpf");
produces<nanoaod::FlatTable>("npf");
produces<nanoaod::FlatTable>("sv");
produces<nanoaod::FlatTable>("svAdapted");
produces<nanoaod::FlatTable>("length");
produces<nanoaod::FlatTable>("muon") ;
produces<nanoaod::FlatTable>("electron") ;
Expand Down Expand Up @@ -76,13 +77,15 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
std::vector<int> cpf_jetIdx;
std::vector<int> npf_jetIdx;
std::vector<int> sv_jetIdx;
std::vector<int> sv_adapted_jetIdx;
std::vector<int> mu_jetIdx;
std::vector<int> elec_jetIdx;

auto lengthTable = std::make_unique<nanoaod::FlatTable>(ntags, "length", false, false);
std::vector<int> cpf_length;
std::vector<int> npf_length;
std::vector<int> sv_length;
std::vector<int> sv_adapted_length;

std::vector<int> elec_length;
std::vector<int> mu_length;
Expand Down Expand Up @@ -167,6 +170,7 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
std::vector<int> cpf_matchedMuon;
std::vector<int> cpf_matchedElectron;
std::vector<int> cpf_matchedSV;
std::vector<int> cpf_matchedSV_adapted;
std::vector<int> cpf_track_numberOfValidPixelHits;
std::vector<int> cpf_track_pixelLayersWithMeasurement;
std::vector<int> cpf_track_numberOfValidStripHits;
Expand Down Expand Up @@ -198,12 +202,26 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
std::vector<float> sv_costhetasvpv;
std::vector<float> sv_enratio;

std::vector<int> mu_isGlobal ;
std::vector<int> mu_isTight ;
std::vector<int> mu_isMedium ;
std::vector<int> mu_isLoose ;
std::vector<int> mu_isStandAlone ;

std::vector<float> sv_adapted_ptrel;
std::vector<float> sv_adapted_deta;
std::vector<float> sv_adapted_dphi;
std::vector<float> sv_adapted_deltaR;
std::vector<float> sv_adapted_mass;
std::vector<float> sv_adapted_ntracks;
std::vector<float> sv_adapted_chi2;
std::vector<int> sv_adapted_ndof;
std::vector<float> sv_adapted_dxy;
std::vector<float> sv_adapted_dxysig;
std::vector<float> sv_adapted_d3d;
std::vector<float> sv_adapted_d3dsig;
std::vector<float> sv_adapted_costhetasvpv;
std::vector<float> sv_adapted_enratio;

std::vector<int> mu_isGlobal;
std::vector<int> mu_isTight;
std::vector<int> mu_isMedium;
std::vector<int> mu_isLoose;
std::vector<int> mu_isStandAlone;

std::vector<float> mu_ptrel;
std::vector<float> mu_EtaRel;
Expand All @@ -227,7 +245,7 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
std::vector<float> mu_dzError;
std::vector<int> mu_numberOfValidPixelHits;
std::vector<int> mu_numberOfpixelLayersWithMeasurement;
std::vector<int> mu_numberOfstripLayersWithMeasurement ;
std::vector<int> mu_numberOfstripLayersWithMeasurement;

std::vector<float> mu_chi2;
std::vector<int> mu_ndof;
Expand All @@ -246,8 +264,6 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
std::vector<float> mu_Pfpileup03;
std::vector<float> mu_sumPfPhotonEt03;



std::vector<float> mu_timeAtIpInOut;
std::vector<float> mu_timeAtIpInOutErr;
std::vector<float> mu_timeAtIpOutIn;
Expand All @@ -261,7 +277,7 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
std::vector<float> elec_dphi;
std::vector<float> elec_charge;
std::vector<float> elec_energy;
std::vector<float> elec_EtFromCaloEn ;
std::vector<float> elec_EtFromCaloEn;
std::vector<float> elec_isEB;
std::vector<float> elec_isEE;
std::vector<float> elec_ecalEnergy;
Expand Down Expand Up @@ -343,6 +359,7 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
unsigned int ncpf_total = 0;
unsigned int nnpf_total = 0;
unsigned int nsv_total = 0;
unsigned int nsv_total_adapted = 0;

for (unsigned int itag= 0; itag < ntags; itag++){
const auto& features = tag_infos->at(itag).features();
Expand All @@ -353,14 +370,19 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
unsigned int ncpf = features.cpf_features.size();
unsigned int nnpf = features.npf_features.size();
unsigned int nsv = features.sv_features.size();
unsigned int nsv_adapted = features.sv_adapted_features.size();

nmu_total += nmu;
nelec_total += nelec;
ncpf_total += ncpf;
nnpf_total += nnpf;
nsv_total += nsv;
nsv_total_adapted += nsv_adapted;

cpf_length.push_back(ncpf);
npf_length.push_back(nnpf);
sv_length.push_back(nsv);
sv_adapted_length.push_back(nsv_adapted);
elec_length.push_back(nelec);
mu_length.push_back(nmu);

Expand Down Expand Up @@ -435,6 +457,7 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
auto cpfTable = std::make_unique<nanoaod::FlatTable>(ncpf_total, "cpf", false, false);
auto npfTable = std::make_unique<nanoaod::FlatTable>(nnpf_total, "npf", false, false);
auto svTable = std::make_unique<nanoaod::FlatTable>(nsv_total, "sv", false, false);
auto svTable_adapted = std::make_unique<nanoaod::FlatTable>(nsv_total_adapted, "svAdapted", false, false);

for (unsigned int itag= 0; itag < ntags; itag++)
{
Expand All @@ -444,12 +467,14 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
auto cpf = features.cpf_features;
auto npf = features.npf_features;
auto sv = features.sv_features;
auto sv_adapted = features.sv_adapted_features;

unsigned int nmu = features.mu_features.size();
unsigned int nelec = features.elec_features.size();
unsigned int ncpf = features.cpf_features.size();
unsigned int nnpf = features.npf_features.size();
unsigned int nsv = features.sv_features.size();
unsigned int nsv_adapted = features.sv_adapted_features.size();

int jetIdx = global_jetIdx[itag];

Expand Down Expand Up @@ -483,6 +508,7 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
cpf_matchedMuon.push_back(cpf_features.cpf_matchedMuon);
cpf_matchedElectron.push_back(cpf_features.cpf_matchedElectron);
cpf_matchedSV.push_back(cpf_features.cpf_matchedSV);
cpf_matchedSV_adapted.push_back(cpf_features.cpf_matchedSV_adapted);
cpf_track_numberOfValidPixelHits.push_back(cpf_features.cpf_track_numberOfValidPixelHits);
cpf_track_pixelLayersWithMeasurement.push_back(cpf_features.cpf_track_pixelLayersWithMeasurement);
cpf_track_numberOfValidStripHits.push_back(cpf_features.cpf_track_numberOfValidStripHits);
Expand Down Expand Up @@ -525,6 +551,26 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
sv_enratio.push_back(sv_features.sv_enratio);
}

for (unsigned int i = 0; i < nsv_adapted; i++)
{
const auto& sv_adapted_features = sv_adapted.at(i);
sv_adapted_jetIdx.push_back(jetIdx);
sv_adapted_ptrel.push_back(sv_adapted_features.sv_ptrel);
sv_adapted_deta.push_back(sv_adapted_features.sv_deta);
sv_adapted_dphi.push_back(sv_adapted_features.sv_dphi);
sv_adapted_deltaR.push_back(sv_adapted_features.sv_deltaR);
sv_adapted_mass.push_back(sv_adapted_features.sv_mass);
sv_adapted_ntracks.push_back(sv_adapted_features.sv_ntracks);
sv_adapted_chi2.push_back(sv_adapted_features.sv_chi2);
sv_adapted_ndof.push_back(sv_adapted_features.sv_ndof);
sv_adapted_dxy.push_back(sv_adapted_features.sv_dxy);
sv_adapted_dxysig.push_back(sv_adapted_features.sv_dxysig);
sv_adapted_d3d.push_back(sv_adapted_features.sv_d3d);
sv_adapted_d3dsig.push_back(sv_adapted_features.sv_d3dsig);
sv_adapted_costhetasvpv.push_back(sv_adapted_features.sv_costhetasvpv);
sv_adapted_enratio.push_back(sv_adapted_features.sv_enratio);
}


for(unsigned int i = 0; i < nmu; i++)
{
Expand Down Expand Up @@ -752,6 +798,7 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
cpfTable->addColumn<int>("matchedMuon", cpf_matchedMuon, "flag to specify whether the track is matched to a PF muon", nanoaod::FlatTable::IntColumn);
cpfTable->addColumn<int>("matchedElectron", cpf_matchedElectron, "flag to specify whether the track is matched to a PF electron", nanoaod::FlatTable::IntColumn);
cpfTable->addColumn<int>("matchedSV", cpf_matchedSV, "flag to specify whether the track is matched to a PF secondary vertex", nanoaod::FlatTable::IntColumn);
cpfTable->addColumn<int>("matchedSV_adapted", cpf_matchedSV_adapted, "flag to specify whether the track is matched to a PF secondary vertex", nanoaod::FlatTable::IntColumn);
cpfTable->addColumn<int>("numberOfValidPixelHits", cpf_track_numberOfValidPixelHits , "number of valid pixel hits " , nanoaod::FlatTable::IntColumn);
cpfTable->addColumn<int>("pixelLayersWithMeasurement", cpf_track_pixelLayersWithMeasurement , "pixel layers with measurment ", nanoaod::FlatTable::IntColumn);
cpfTable->addColumn<int>("numberOfValidStripHits" , cpf_track_numberOfValidStripHits , "nb of valid strip hits " , nanoaod::FlatTable::IntColumn);
Expand Down Expand Up @@ -788,6 +835,22 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
svTable->addColumn<float>("costhetasvpv", sv_costhetasvpv, "cos(theta) of the secondary vertex (SV) with respect to the primary vertex (PV)", nanoaod::FlatTable::FloatColumn);
svTable->addColumn<float>("enratio", sv_enratio, "ratio of secondary vertex (SV) energy with respect to the jet", nanoaod::FlatTable::FloatColumn);

svTable_adapted->addColumn<int>("jetIdx", sv_adapted_jetIdx, "linked jet Id", nanoaod::FlatTable::IntColumn);
svTable_adapted->addColumn<float>("ptrel", sv_adapted_ptrel, "secondary vertex (SV) transverse momentum relative to the uncorrected jet energy pt", nanoaod::FlatTable::FloatColumn);
svTable_adapted->addColumn<float>("deta", sv_adapted_deta, "absolute difference between the secondary vertex (SV) eta and jet eta", nanoaod::FlatTable::FloatColumn);
svTable_adapted->addColumn<float>("dphi", sv_adapted_dphi, "absolute difference between the secondary vertex (SV) phi and jet phi", nanoaod::FlatTable::FloatColumn);
svTable_adapted->addColumn<float>("deltaR", sv_adapted_deltaR, "pseudoangular distance (Delta R) between jet axis and SV flight direction", nanoaod::FlatTable::FloatColumn);
svTable_adapted->addColumn<float>("mass", sv_adapted_mass, "invariant mass of reconstructed secondary vertex (SV)", nanoaod::FlatTable::FloatColumn);
svTable_adapted->addColumn<float>("ntracks", sv_adapted_ntracks, "number of tracks associated to the secondary vertex (SV)", nanoaod::FlatTable::FloatColumn);
svTable_adapted->addColumn<float>("chi2", sv_adapted_chi2, "chi^2 of secondary vertex (SV)", nanoaod::FlatTable::FloatColumn);
svTable_adapted->addColumn<int>("ndof", sv_adapted_ndof, "number of degrees of freedom of secondary vertex (SV) fit", nanoaod::FlatTable::IntColumn);
svTable_adapted->addColumn<float>("dxy", sv_adapted_dxy, "transverse impact parameter of the secondary vertex (SV)", nanoaod::FlatTable::FloatColumn);
svTable_adapted->addColumn<float>("dxysig", sv_adapted_dxysig, "transverse impact parameter significance of the secondary vertex (SV)", nanoaod::FlatTable::FloatColumn);
svTable_adapted->addColumn<float>("d3d", sv_adapted_d3d, "3D impact parameter of secondary vertex (SV)", nanoaod::FlatTable::FloatColumn);
svTable_adapted->addColumn<float>("d3dsig", sv_adapted_d3dsig, "3D impact parameter significance of secondary vertex (SV)", nanoaod::FlatTable::FloatColumn);
svTable_adapted->addColumn<float>("costhetasvpv", sv_adapted_costhetasvpv, "cos(theta) of the secondary vertex (SV) with respect to the primary vertex (PV)", nanoaod::FlatTable::FloatColumn);
svTable_adapted->addColumn<float>("enratio", sv_adapted_enratio, "ratio of secondary vertex (SV) energy with respect to the jet", nanoaod::FlatTable::FloatColumn);

lengthTable->addColumn<int>("cpf", cpf_length, "charged PF candidate track offset", nanoaod::FlatTable::IntColumn);
lengthTable->addColumn<int>("npf", npf_length, "neutral PF candidate offset", nanoaod::FlatTable::IntColumn);
lengthTable->addColumn<int>("sv", sv_length, "secondary vertex (SV) offset", nanoaod::FlatTable::IntColumn);
Expand Down Expand Up @@ -939,6 +1002,7 @@ NANOProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSetup)
iEvent.put(std::move(cpfTable), "cpf");
iEvent.put(std::move(npfTable), "npf");
iEvent.put(std::move(svTable), "sv");
iEvent.put(std::move(svTable_adapted), "svAdapted");
iEvent.put(std::move(lengthTable), "length");
iEvent.put(std::move(muonTable), "muon");
iEvent.put(std::move(electronTable), "electron");
Expand Down
14 changes: 8 additions & 6 deletions NANOProducer/python/adaptedSV_cff.py
View file
Original file line number Diff line number Diff line change
@@ -1,5 +1,7 @@
import FWCore.ParameterSet.Config as cms

# https://gist.github.com/daiktas/d3e0ed47b82e45502aaf787ae12934fd

adaptedInclusiveCandidateVertexFinder = cms.EDProducer("InclusiveCandidateVertexFinder",
beamSpot = cms.InputTag("offlineBeamSpot"),
clusterizer = cms.PSet(
Expand All @@ -16,8 +18,8 @@
fitterSigmacut = cms.double(3),
fitterTini = cms.double(256),
maxNTracks = cms.uint32(30),
maximumLongitudinalImpactParameter = cms.double(3), # Default is 0.3->3
minHits = cms.uint32(0),
maximumLongitudinalImpactParameter = cms.double(1), # Default is 0.3->20 (Ghent) -> 1 may be safer
minHits = cms.uint32(6), # 8 -> 6
minPt = cms.double(0.8),
primaryVertices = cms.InputTag("offlineSlimmedPrimaryVertices"),
tracks = cms.InputTag("packedPFCandidates"),
Expand All @@ -43,17 +45,17 @@
adaptedCandidateVertexArbitrator = cms.EDProducer("CandidateVertexArbitrator",
beamSpot = cms.InputTag("offlineBeamSpot"),
dLenFraction = cms.double(0.333),
dRCut = cms.double(0.4),
distCut = cms.double(0.04),
dRCut = cms.double(0.4), # 0.4 -> 1 # not sure good idea...
distCut = cms.double(0.04), # 0.04 -> 0.1 # check if safe
fitterRatio = cms.double(0.25),
fitterSigmacut = cms.double(3),
fitterTini = cms.double(256),
primaryVertices = cms.InputTag("offlineSlimmedPrimaryVertices"),
secondaryVertices = cms.InputTag("adaptedCandidateVertexMerger"),
sigCut = cms.double(5),
trackMinLayers = cms.int32(4),
trackMinPixels = cms.int32(0), # Default is 1->0
trackMinPt = cms.double(0.4),
trackMinPixels = cms.int32(0), # 1->0
trackMinPt = cms.double(0.8),
tracks = cms.InputTag("packedPFCandidates")
)

Expand Down
2 changes: 1 addition & 1 deletion NANOProducer/test/produceNANO.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -224,7 +224,7 @@
electronSrc = cms.InputTag("slimmedElectrons"),
shallow_tag_infos = cms.InputTag('pfDeepCSVTagInfosXTag'),
vertices = cms.InputTag('offlineSlimmedPrimaryVertices'),
#secondary_vertices = cms.InputTag("adaptedSlimmedSecondaryVertices")
secondary_vertices_adapted = cms.InputTag("adaptedSlimmedSecondaryVertices"),
secondary_vertices = cms.InputTag("slimmedSecondaryVertices")
)

Expand Down
Loading

0 comments on commit 337515d

Please sign in to comment.