def SS_selection(lep1, lep2):
selection = PackedSelection()
is_dilep = ((ak.num(lep1) + ak.num(lep2))==2)
pos_charge = ((ak.sum(lep1.pdgId, axis=1) + ak.sum(lep2.pdgId, axis=1))<0)
neg_charge = ((ak.sum(lep1.pdgId, axis=1) + ak.sum(lep2.pdgId, axis=1))>0)
dilep2 = choose(lep2, 2)
dilep1 = choose(lep1, 2)
dilep = cross(lep2, lep1)
is_SS = ( ak.any((dilep2['0'].charge * dilep2['1'].charge)>0, axis=1) | \
ak.any((dilep1['0'].charge * dilep1['1'].charge)>0, axis=1) | \
ak.any((dilep['0'].charge * dilep['1'].charge)>0, axis=1) )
selection.add('SS', is_SS)
ss_reqs = ['SS']
ss_reqs_d = {sel: True for sel in ss_reqs}
ss_selection = selection.require(**ss_reqs_d)
return ss_selection
def test_packed_selection():
from coffea.analysis_tools import PackedSelection
sel = PackedSelection()
counts, test_eta, test_pt = dummy_jagged_eta_pt()
all_true = np.full(shape=counts.shape, fill_value=True, dtype=np.bool)
all_false = np.full(shape=counts.shape, fill_value=False, dtype=np.bool)
ones = np.ones(shape=counts.shape, dtype=np.uint64)
wrong_shape = ones = np.ones(shape=(counts.shape[0] - 5, ), dtype=np.bool)
sel.add("all_true", all_true)
sel.add("all_false", all_false)
assert np.all(sel.require(all_true=True, all_false=False) == all_true)
assert np.all(sel.all("all_true", "all_false") == all_false)
try:
sel.require(all_true=1, all_false=0)
except ValueError:
pass
try:
sel.add("wrong_shape", wrong_shape)
except ValueError:
pass
try:
sel.add("ones", ones)
except ValueError:
pass
try:
overpack = PackedSelection()
for i in range(65):
overpack.add("sel_%d", all_true)
except RuntimeError:
pass
def process(self, events):
# Dataset parameters
dataset = events.metadata['dataset']
histAxisName = self._samples[dataset]['histAxisName']
year = self._samples[dataset]['year']
xsec = self._samples[dataset]['xsec']
sow = self._samples[dataset]['nSumOfWeights']
isData = self._samples[dataset]['isData']
datasets = [
'SingleMuon', 'SingleElectron', 'EGamma', 'MuonEG', 'DoubleMuon',
'DoubleElectron'
]
for d in datasets:
if d in dataset: dataset = dataset.split('_')[0]
# Initialize objects
met = events.MET
e = events.Electron
mu = events.Muon
tau = events.Tau
j = events.Jet
# Muon selection
mu['isPres'] = isPresMuon(mu.dxy, mu.dz, mu.sip3d, mu.looseId)
mu['isTight'] = isTightMuon(mu.pt,
mu.eta,
mu.dxy,
mu.dz,
mu.pfRelIso03_all,
mu.sip3d,
mu.mvaTTH,
mu.mediumPromptId,
mu.tightCharge,
mu.looseId,
minpt=10)
mu['isGood'] = mu['isPres'] & mu['isTight']
leading_mu = mu[ak.argmax(mu.pt, axis=-1, keepdims=True)]
leading_mu = leading_mu[leading_mu.isGood]
mu = mu[mu.isGood]
mu_pres = mu[mu.isPres]
# Electron selection
e['isPres'] = isPresElec(e.pt,
e.eta,
e.dxy,
e.dz,
e.miniPFRelIso_all,
e.sip3d,
e.lostHits,
minpt=15)
e['isTight'] = isTightElec(e.pt,
e.eta,
e.dxy,
e.dz,
e.miniPFRelIso_all,
e.sip3d,
e.mvaTTH,
e.mvaFall17V2Iso,
e.lostHits,
e.convVeto,
e.tightCharge,
e.sieie,
e.hoe,
e.eInvMinusPInv,
minpt=15)
e['isClean'] = isClean(e, mu, drmin=0.05)
e['isGood'] = e['isPres'] & e['isTight'] & e['isClean']
leading_e = e[ak.argmax(e.pt, axis=-1, keepdims=True)]
leading_e = leading_e[leading_e.isGood]
e = e[e.isGood]
e_pres = e[e.isPres & e.isClean]
# Tau selection
tau['isPres'] = isPresTau(tau.pt,
tau.eta,
tau.dxy,
tau.dz,
tau.leadTkPtOverTauPt,
tau.idAntiMu,
tau.idAntiEle,
tau.rawIso,
tau.idDecayModeNewDMs,
minpt=20)
tau['isClean'] = isClean(tau, e_pres, drmin=0.4) & isClean(
tau, mu_pres, drmin=0.4)
tau['isGood'] = tau['isPres'] # & tau['isClean'], for the moment
tau = tau[tau.isGood]
nElec = ak.num(e)
nMuon = ak.num(mu)
nTau = ak.num(tau)
twoLeps = (nElec + nMuon) == 2
threeLeps = (nElec + nMuon) == 3
twoElec = (nElec == 2)
twoMuon = (nMuon == 2)
e0 = e[ak.argmax(e.pt, axis=-1, keepdims=True)]
m0 = mu[ak.argmax(mu.pt, axis=-1, keepdims=True)]
# Attach the lepton SFs to the electron and muons collections
AttachElectronSF(e, year=year)
AttachMuonSF(mu, year=year)
# Create a lepton (muon+electron) collection and calculate a per event lepton SF
leps = ak.concatenate([e, mu], axis=-1)
events['lepSF_nom'] = ak.prod(leps.sf_nom, axis=-1)
events['lepSF_hi'] = ak.prod(leps.sf_hi, axis=-1)
events['lepSF_lo'] = ak.prod(leps.sf_lo, axis=-1)
# Jet selection
jetptname = 'pt_nom' if hasattr(j, 'pt_nom') else 'pt'
### Jet energy corrections
if not isData:
j["pt_raw"] = (1 - j.rawFactor) * j.pt
j["mass_raw"] = (1 - j.rawFactor) * j.mass
j["pt_gen"] = ak.values_astype(ak.fill_none(j.matched_gen.pt, 0),
np.float32)
j["rho"] = ak.broadcast_arrays(events.fixedGridRhoFastjetAll,
j.pt)[0]
events_cache = events.caches[0]
corrected_jets = jet_factory.build(j, lazy_cache=events_cache)
#print('jet pt: ',j.pt)
#print('cor pt: ',corrected_jets.pt)
#print('jes up: ',corrected_jets.JES_jes.up.pt)
#print('jes down: ',corrected_jets.JES_jes.down.pt)
#print(ak.fields(corrected_jets))
'''
# SYSTEMATICS
jets = corrected_jets
if(self.jetSyst == 'JERUp'):
jets = corrected_jets.JER.up
elif(self.jetSyst == 'JERDown'):
jets = corrected_jets.JER.down
elif(self.jetSyst == 'JESUp'):
jets = corrected_jets.JES_jes.up
elif(self.jetSyst == 'JESDown'):
jets = corrected_jets.JES_jes.down
'''
j['isGood'] = isTightJet(getattr(j,
jetptname), j.eta, j.jetId, j.neHEF,
j.neEmEF, j.chHEF, j.chEmEF, j.nConstituents)
#j['isgood'] = isGoodJet(j.pt, j.eta, j.jetId)
#j['isclean'] = isClean(j, e, mu)
j['isClean'] = isClean(j, e, drmin=0.4) & isClean(
j, mu, drmin=0.4) # & isClean(j, tau, drmin=0.4)
goodJets = j[(j.isClean) & (j.isGood)]
njets = ak.num(goodJets)
ht = ak.sum(goodJets.pt, axis=-1)
j0 = goodJets[ak.argmax(goodJets.pt, axis=-1, keepdims=True)]
#nbtags = ak.num(goodJets[goodJets.btagDeepFlavB > 0.2770])
# Loose DeepJet WP
if year == 2017: btagwpl = 0.0532 #WP loose
else: btagwpl = 0.0490 #WP loose
isBtagJetsLoose = (goodJets.btagDeepB > btagwpl)
isNotBtagJetsLoose = np.invert(isBtagJetsLoose)
nbtagsl = ak.num(goodJets[isBtagJetsLoose])
# Medium DeepJet WP
if year == 2017: btagwpm = 0.3040 #WP medium
else: btagwpm = 0.2783 #WP medium
isBtagJetsMedium = (goodJets.btagDeepB > btagwpm)
isNotBtagJetsMedium = np.invert(isBtagJetsMedium)
nbtagsm = ak.num(goodJets[isBtagJetsMedium])
# Btag SF following 1a) in https://twiki.cern.ch/twiki/bin/viewauth/CMS/BTagSFMethods
btagSF = np.ones_like(ht)
btagSFUp = np.ones_like(ht)
btagSFDo = np.ones_like(ht)
if not isData:
pt = goodJets.pt
abseta = np.abs(goodJets.eta)
flav = goodJets.hadronFlavour
bJetSF = GetBTagSF(abseta, pt, flav)
bJetSFUp = GetBTagSF(abseta, pt, flav, sys=1)
bJetSFDo = GetBTagSF(abseta, pt, flav, sys=-1)
bJetEff = GetBtagEff(abseta, pt, flav, year)
bJetEff_data = bJetEff * bJetSF
bJetEff_dataUp = bJetEff * bJetSFUp
bJetEff_dataDo = bJetEff * bJetSFDo
pMC = ak.prod(bJetEff[isBtagJetsMedium], axis=-1) * ak.prod(
(1 - bJetEff[isNotBtagJetsMedium]), axis=-1)
pData = ak.prod(bJetEff_data[isBtagJetsMedium], axis=-1) * ak.prod(
(1 - bJetEff_data[isNotBtagJetsMedium]), axis=-1)
pDataUp = ak.prod(
bJetEff_dataUp[isBtagJetsMedium], axis=-1) * ak.prod(
(1 - bJetEff_dataUp[isNotBtagJetsMedium]), axis=-1)
pDataDo = ak.prod(
bJetEff_dataDo[isBtagJetsMedium], axis=-1) * ak.prod(
(1 - bJetEff_dataDo[isNotBtagJetsMedium]), axis=-1)
pMC = ak.where(pMC == 0, 1,
pMC) # removeing zeroes from denominator...
btagSF = pData / pMC
btagSFUp = pDataUp / pMC
btagSFDo = pDataUp / pMC
##################################################################
### 2 same-sign leptons
##################################################################
# emu
singe = e[(nElec == 1) & (nMuon == 1) & (e.pt > -1)]
singm = mu[(nElec == 1) & (nMuon == 1) & (mu.pt > -1)]
em = ak.cartesian({"e": singe, "m": singm})
emSSmask = (em.e.charge * em.m.charge > 0)
emSS = em[emSSmask]
nemSS = len(ak.flatten(emSS))
# ee and mumu
# pt>-1 to preserve jagged dimensions
ee = e[(nElec == 2) & (nMuon == 0) & (e.pt > -1)]
mm = mu[(nElec == 0) & (nMuon == 2) & (mu.pt > -1)]
sumcharge = ak.sum(e.charge, axis=-1) + ak.sum(mu.charge, axis=-1)
eepairs = ak.combinations(ee, 2, fields=["e0", "e1"])
eeSSmask = (eepairs.e0.charge * eepairs.e1.charge > 0)
eeonZmask = (np.abs((eepairs.e0 + eepairs.e1).mass - 91.2) < 10)
eeoffZmask = (eeonZmask == 0)
mmpairs = ak.combinations(mm, 2, fields=["m0", "m1"])
mmSSmask = (mmpairs.m0.charge * mmpairs.m1.charge > 0)
mmonZmask = (np.abs((mmpairs.m0 + mmpairs.m1).mass - 91.2) < 10)
mmoffZmask = (mmonZmask == 0)
eeSSonZ = eepairs[eeSSmask & eeonZmask]
eeSSoffZ = eepairs[eeSSmask & eeoffZmask]
mmSSonZ = mmpairs[mmSSmask & mmonZmask]
mmSSoffZ = mmpairs[mmSSmask & mmoffZmask]
neeSS = len(ak.flatten(eeSSonZ)) + len(ak.flatten(eeSSoffZ))
nmmSS = len(ak.flatten(mmSSonZ)) + len(ak.flatten(mmSSoffZ))
print('Same-sign events [ee, emu, mumu] = [%i, %i, %i]' %
(neeSS, nemSS, nmmSS))
# Cuts
eeSSmask = (ak.num(eeSSmask[eeSSmask]) > 0)
mmSSmask = (ak.num(mmSSmask[mmSSmask]) > 0)
eeonZmask = (ak.num(eeonZmask[eeonZmask]) > 0)
eeoffZmask = (ak.num(eeoffZmask[eeoffZmask]) > 0)
mmonZmask = (ak.num(mmonZmask[mmonZmask]) > 0)
mmoffZmask = (ak.num(mmoffZmask[mmoffZmask]) > 0)
emSSmask = (ak.num(emSSmask[emSSmask]) > 0)
##################################################################
### 3 leptons
##################################################################
# eem
muon_eem = mu[(nElec == 2) & (nMuon == 1) & (mu.pt > -1)]
elec_eem = e[(nElec == 2) & (nMuon == 1) & (e.pt > -1)]
ee_eem = ak.combinations(elec_eem, 2, fields=["e0", "e1"])
ee_eemZmask = (ee_eem.e0.charge * ee_eem.e1.charge < 1) & (np.abs(
(ee_eem.e0 + ee_eem.e1).mass - 91.2) < 10)
ee_eemOffZmask = (ee_eem.e0.charge * ee_eem.e1.charge < 1) & (np.abs(
(ee_eem.e0 + ee_eem.e1).mass - 91.2) > 10)
ee_eemZmask = (ak.num(ee_eemZmask[ee_eemZmask]) > 0)
ee_eemOffZmask = (ak.num(ee_eemOffZmask[ee_eemOffZmask]) > 0)
eepair_eem = (ee_eem.e0 + ee_eem.e1)
trilep_eem = eepair_eem + muon_eem #ak.cartesian({"e0":ee_eem.e0,"e1":ee_eem.e1, "m":muon_eem})
# mme
muon_mme = mu[(nElec == 1) & (nMuon == 2) & (mu.pt > -1)]
elec_mme = e[(nElec == 1) & (nMuon == 2) & (e.pt > -1)]
mm_mme = ak.combinations(muon_mme, 2, fields=["m0", "m1"])
mm_mmeZmask = (mm_mme.m0.charge * mm_mme.m1.charge < 1) & (np.abs(
(mm_mme.m0 + mm_mme.m1).mass - 91.2) < 10)
mm_mmeOffZmask = (mm_mme.m0.charge * mm_mme.m1.charge < 1) & (np.abs(
(mm_mme.m0 + mm_mme.m1).mass - 91.2) > 10)
mm_mmeZmask = (ak.num(mm_mmeZmask[mm_mmeZmask]) > 0)
mm_mmeOffZmask = (ak.num(mm_mmeOffZmask[mm_mmeOffZmask]) > 0)
mmpair_mme = (mm_mme.m0 + mm_mme.m1)
trilep_mme = mmpair_mme + elec_mme
mZ_mme = mmpair_mme.mass
mZ_eem = eepair_eem.mass
m3l_eem = trilep_eem.mass
m3l_mme = trilep_mme.mass
# eee and mmm
eee = e[(nElec == 3) & (nMuon == 0) & (e.pt > -1)]
mmm = mu[(nElec == 0) & (nMuon == 3) & (mu.pt > -1)]
eee_leps = ak.combinations(eee, 3, fields=["e0", "e1", "e2"])
mmm_leps = ak.combinations(mmm, 3, fields=["m0", "m1", "m2"])
ee_pairs = ak.combinations(eee, 2, fields=["e0", "e1"])
mm_pairs = ak.combinations(mmm, 2, fields=["m0", "m1"])
ee_pairs_index = ak.argcombinations(eee, 2, fields=["e0", "e1"])
mm_pairs_index = ak.argcombinations(mmm, 2, fields=["m0", "m1"])
mmSFOS_pairs = mm_pairs[
(np.abs(mm_pairs.m0.pdgId) == np.abs(mm_pairs.m1.pdgId))
& (mm_pairs.m0.charge != mm_pairs.m1.charge)]
offZmask_mm = ak.all(
np.abs((mmSFOS_pairs.m0 + mmSFOS_pairs.m1).mass - 91.2) > 10.,
axis=1,
keepdims=True) & (ak.num(mmSFOS_pairs) > 0)
onZmask_mm = ak.any(
np.abs((mmSFOS_pairs.m0 + mmSFOS_pairs.m1).mass - 91.2) < 10.,
axis=1,
keepdims=True)
eeSFOS_pairs = ee_pairs[
(np.abs(ee_pairs.e0.pdgId) == np.abs(ee_pairs.e1.pdgId))
& (ee_pairs.e0.charge != ee_pairs.e1.charge)]
offZmask_ee = ak.all(
np.abs((eeSFOS_pairs.e0 + eeSFOS_pairs.e1).mass - 91.2) > 10,
axis=1,
keepdims=True) & (ak.num(eeSFOS_pairs) > 0)
onZmask_ee = ak.any(
np.abs((eeSFOS_pairs.e0 + eeSFOS_pairs.e1).mass - 91.2) < 10,
axis=1,
keepdims=True)
# Create masks **for event selection**
eeeOnZmask = (ak.num(onZmask_ee[onZmask_ee]) > 0)
eeeOffZmask = (ak.num(offZmask_ee[offZmask_ee]) > 0)
mmmOnZmask = (ak.num(onZmask_mm[onZmask_mm]) > 0)
mmmOffZmask = (ak.num(offZmask_mm[offZmask_mm]) > 0)
# Now we need to create masks for the leptons in order to select leptons from the Z boson candidate (in onZ categories)
ZeeMask = ak.argmin(np.abs((eeSFOS_pairs.e0 + eeSFOS_pairs.e1).mass -
91.2),
axis=1,
keepdims=True)
ZmmMask = ak.argmin(np.abs((mmSFOS_pairs.m0 + mmSFOS_pairs.m1).mass -
91.2),
axis=1,
keepdims=True)
Zee = eeSFOS_pairs[ZeeMask]
Zmm = mmSFOS_pairs[ZmmMask]
eZ0 = Zee.e0[ak.num(eeSFOS_pairs) > 0]
eZ1 = Zee.e1[ak.num(eeSFOS_pairs) > 0]
eZ = eZ0 + eZ1
mZ0 = Zmm.m0[ak.num(mmSFOS_pairs) > 0]
mZ1 = Zmm.m1[ak.num(mmSFOS_pairs) > 0]
mZ = mZ0 + mZ1
mZ_eee = eZ.mass
mZ_mmm = mZ.mass
# And for the W boson
ZmmIndices = mm_pairs_index[ZmmMask]
ZeeIndices = ee_pairs_index[ZeeMask]
eW = eee[~ZeeIndices.e0 | ~ZeeIndices.e1]
mW = mmm[~ZmmIndices.m0 | ~ZmmIndices.m1]
triElec = eee_leps.e0 + eee_leps.e1 + eee_leps.e2
triMuon = mmm_leps.m0 + mmm_leps.m1 + mmm_leps.m2
m3l_eee = triElec.mass
m3l_mmm = triMuon.mass
##################################################################
### >=4 leptons
##################################################################
# 4lep cat
is4lmask = ((nElec + nMuon) >= 4)
muon_4l = mu[(is4lmask) & (mu.pt > -1)]
elec_4l = e[(is4lmask) & (e.pt > -1)]
# selecting 4 leading leptons
leptons = ak.concatenate([e, mu], axis=-1)
leptons_sorted = leptons[ak.argsort(leptons.pt,
axis=-1,
ascending=False)]
lep4l = leptons_sorted[:, 0:4]
e4l = lep4l[abs(lep4l.pdgId) == 11]
mu4l = lep4l[abs(lep4l.pdgId) == 13]
nElec4l = ak.num(e4l)
nMuon4l = ak.num(mu4l)
# Triggers
trig_eeSS = passTrigger(events, 'ee', isData, dataset)
trig_mmSS = passTrigger(events, 'mm', isData, dataset)
trig_emSS = passTrigger(events, 'em', isData, dataset)
trig_eee = passTrigger(events, 'eee', isData, dataset)
trig_mmm = passTrigger(events, 'mmm', isData, dataset)
trig_eem = passTrigger(events, 'eem', isData, dataset)
trig_mme = passTrigger(events, 'mme', isData, dataset)
trig_4l = triggerFor4l(events, nMuon, nElec, isData, dataset)
# MET filters
# Weights
genw = np.ones_like(events['event']) if (
isData or len(self._wc_names_lst) > 0) else events['genWeight']
### We need weights for: normalization, lepSF, triggerSF, pileup, btagSF...
weights = {}
for r in [
'all', 'ee', 'mm', 'em', 'eee', 'mmm', 'eem', 'mme', 'eeee',
'eeem', 'eemm', 'mmme', 'mmmm'
]:
# weights[r] = coffea.analysis_tools.Weights(len(events))
weights[r] = coffea.analysis_tools.Weights(len(events),
storeIndividual=True)
if len(self._wc_names_lst) > 0:
sow = np.ones_like(
sow
) # Not valid in nanoAOD for EFT samples, MUST use SumOfEFTweights at analysis level
weights[r].add('norm', genw if isData else (xsec / sow) * genw)
weights[r].add('btagSF', btagSF, btagSFUp, btagSFDo)
weights[r].add('lepSF', events.lepSF_nom, events.lepSF_hi,
events.lepSF_lo)
# Extract the EFT quadratic coefficients and optionally use them to calculate the coefficients on the w**2 quartic function
# eft_coeffs is never Jagged so convert immediately to numpy for ease of use.
eft_coeffs = ak.to_numpy(events['EFTfitCoefficients']) if hasattr(
events, "EFTfitCoefficients") else None
if eft_coeffs is not None:
# Check to see if the ordering of WCs for this sample matches what want
if self._samples[dataset]['WCnames'] != self._wc_names_lst:
eft_coeffs = efth.remap_coeffs(
self._samples[dataset]['WCnames'], self._wc_names_lst,
eft_coeffs)
eft_w2_coeffs = efth.calc_w2_coeffs(eft_coeffs, self._dtype) if (
self._do_errors and eft_coeffs is not None) else None
# Selections and cuts
selections = PackedSelection() #(dtype='uint64')
channels2LSS = ['eeSSonZ', 'eeSSoffZ', 'mmSSonZ', 'mmSSoffZ', 'emSS']
selections.add('eeSSonZ', (eeonZmask) & (eeSSmask) & (trig_eeSS))
selections.add('eeSSoffZ', (eeoffZmask) & (eeSSmask) & (trig_eeSS))
selections.add('mmSSonZ', (mmonZmask) & (mmSSmask) & (trig_mmSS))
selections.add('mmSSoffZ', (mmoffZmask) & (mmSSmask) & (trig_mmSS))
selections.add('emSS', (emSSmask) & (trig_emSS))
channels3L = ['eemSSonZ', 'eemSSoffZ', 'mmeSSonZ', 'mmeSSoffZ']
selections.add('eemSSonZ', (ee_eemZmask) & (trig_eem))
selections.add('eemSSoffZ', (ee_eemOffZmask) & (trig_eem))
selections.add('mmeSSonZ', (mm_mmeZmask) & (trig_mme))
selections.add('mmeSSoffZ', (mm_mmeOffZmask) & (trig_mme))
channels3L += ['eeeSSonZ', 'eeeSSoffZ', 'mmmSSonZ', 'mmmSSoffZ']
selections.add('eeeSSonZ', (eeeOnZmask) & (trig_eee))
selections.add('eeeSSoffZ', (eeeOffZmask) & (trig_eee))
selections.add('mmmSSonZ', (mmmOnZmask) & (trig_mmm))
selections.add('mmmSSoffZ', (mmmOffZmask) & (trig_mmm))
channels4L = ['eeee', 'eeem', 'eemm', 'mmme', 'mmmm']
selections.add('eeee', ((nElec4l == 4) & (nMuon4l == 0)) & (trig_4l))
selections.add('eeem', ((nElec4l == 3) & (nMuon4l == 1)) & (trig_4l))
selections.add('eemm', ((nElec4l == 2) & (nMuon4l == 2)) & (trig_4l))
selections.add('mmme', ((nElec4l == 1) & (nMuon4l == 3)) & (trig_4l))
selections.add('mmmm', ((nElec4l == 0) & (nMuon4l == 4)) & (trig_4l))
selections.add('ch+', (sumcharge > 0))
selections.add('ch-', (sumcharge < 0))
selections.add('ch0', (sumcharge == 0))
levels = ['base', '1+bm2+bl', '1bm', '2+bm']
selections.add('base', (nElec + nMuon >= 2))
selections.add('1+bm2+bl', (nElec + nMuon >= 2) & ((nbtagsm >= 1) &
(nbtagsl >= 2)))
selections.add('1bm', (nElec + nMuon >= 2) & (nbtagsm == 1))
selections.add('2+bm', (nElec + nMuon >= 2) & (nbtagsm >= 2))
# Variables
invMass_eeSSonZ = (eeSSonZ.e0 + eeSSonZ.e1).mass
invMass_eeSSoffZ = (eeSSoffZ.e0 + eeSSoffZ.e1).mass
invMass_mmSSonZ = (mmSSonZ.m0 + mmSSonZ.m1).mass
invMass_mmSSoffZ = (mmSSoffZ.m0 + mmSSoffZ.m1).mass
invMass_emSS = (emSS.e + emSS.m).mass
varnames = {}
varnames['met'] = met.pt
varnames['ht'] = ht
varnames['njets'] = njets
varnames['invmass'] = {
'eeSSonZ': invMass_eeSSonZ,
'eeSSoffZ': invMass_eeSSoffZ,
'mmSSonZ': invMass_mmSSonZ,
'mmSSoffZ': invMass_mmSSoffZ,
'emSS': invMass_emSS,
'eemSSonZ': mZ_eem,
'eemSSoffZ': mZ_eem,
'mmeSSonZ': mZ_mme,
'mmeSSoffZ': mZ_mme,
'eeeSSonZ': mZ_eee,
'eeeSSoffZ': mZ_eee,
'mmmSSonZ': mZ_mmm,
'mmmSSoffZ': mZ_mmm,
}
varnames['m3l'] = {
'eemSSonZ': m3l_eem,
'eemSSoffZ': m3l_eem,
'mmeSSonZ': m3l_mme,
'mmeSSoffZ': m3l_mme,
'eeeSSonZ': m3l_eee,
'eeeSSoffZ': m3l_eee,
'mmmSSonZ': m3l_mmm,
'mmmSSoffZ': m3l_mmm,
}
varnames['e0pt'] = e0.pt
varnames['e0eta'] = e0.eta
varnames['m0pt'] = m0.pt
varnames['m0eta'] = m0.eta
varnames['j0pt'] = j0.pt
varnames['j0eta'] = j0.eta
varnames['counts'] = np.ones_like(events['event'])
# systematics
systList = []
if isData == False:
systList = ['nominal']
if self._do_systematics:
systList = systList + [
'lepSFUp', 'lepSFDown', 'btagSFUp', 'btagSFDown'
]
else:
systList = ['noweight']
# fill Histos
hout = self.accumulator.identity()
normweights = weights['all'].weight().flatten(
) # Why does it not complain about .flatten() here?
sowweights = np.ones_like(normweights) if len(
self._wc_names_lst) > 0 else normweights
hout['SumOfEFTweights'].fill(sample=histAxisName,
SumOfEFTweights=varnames['counts'],
weight=sowweights,
eft_coeff=eft_coeffs,
eft_err_coeff=eft_w2_coeffs)
for syst in systList:
for var, v in varnames.items():
for ch in channels2LSS + channels3L + channels4L:
for sumcharge in ['ch+', 'ch-', 'ch0']:
for lev in levels:
#find the event weight to be used when filling the histograms
weightSyst = syst
#in the case of 'nominal', or the jet energy systematics, no weight systematic variation is used (weightSyst=None)
if syst in [
'nominal', 'JERUp', 'JERDown', 'JESUp',
'JESDown'
]:
weightSyst = None # no weight systematic for these variations
if syst == 'noweight':
weight = np.ones(len(events)) # for data
else:
# call weights.weight() with the name of the systematic to be varied
if ch in channels3L: ch_w = ch[:3]
elif ch in channels2LSS: ch_w = ch[:2]
else: ch_w = ch
weight = weights['all'].weight(
weightSyst
) if isData else weights[ch_w].weight(
weightSyst)
cuts = [ch] + [lev] + [sumcharge]
cut = selections.all(*cuts)
weights_flat = weight[cut].flatten(
) # Why does it not complain about .flatten() here?
weights_ones = np.ones_like(weights_flat,
dtype=np.int)
eft_coeffs_cut = eft_coeffs[
cut] if eft_coeffs is not None else None
eft_w2_coeffs_cut = eft_w2_coeffs[
cut] if eft_w2_coeffs is not None else None
# filling histos
if var == 'invmass':
if ((ch in [
'eeeSSoffZ', 'mmmSSoffZ', 'eeeSSonZ',
'mmmSSonZ'
]) or (ch in channels4L)):
continue
else:
values = ak.flatten(v[ch][cut])
hout['invmass'].fill(
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
invmass=values,
weight=weights_flat,
systematic=syst)
elif var == 'm3l':
if ((ch in channels2LSS) or (ch in [
'eeeSSoffZ', 'mmmSSoffZ', 'eeeSSonZ',
'mmmSSonZ'
]) or (ch in channels4L)):
continue
values = ak.flatten(v[ch][cut])
hout['m3l'].fill(
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
m3l=values,
weight=weights_flat,
systematic=syst)
else:
values = v[cut]
# These all look identical, do we need if/else here?
if var == 'ht':
hout[var].fill(
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut,
ht=values,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
weight=weights_flat,
systematic=syst)
elif var == 'met':
hout[var].fill(
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut,
met=values,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
weight=weights_flat,
systematic=syst)
elif var == 'njets':
hout[var].fill(
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut,
njets=values,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
weight=weights_flat,
systematic=syst)
elif var == 'nbtags':
hout[var].fill(
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut,
nbtags=values,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
weight=weights_flat,
systematic=syst)
elif var == 'counts':
hout[var].fill(counts=values,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
weight=weights_ones,
systematic=syst)
elif var == 'j0eta':
if lev == 'base': continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut,
j0eta=values,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
weight=weights_flat,
systematic=syst)
elif var == 'e0pt':
if ch in [
'mmSSonZ', 'mmSSoffZ', 'mmmSSoffZ',
'mmmSSonZ', 'mmmm'
]:
continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut,
e0pt=values,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
weight=weights_flat,
systematic=syst
) # Crashing here, not sure why. Related to values?
elif var == 'm0pt':
if ch in [
'eeSSonZ', 'eeSSoffZ', 'eeeSSoffZ',
'eeeSSonZ', 'eeee'
]:
continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut,
m0pt=values,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
weight=weights_flat,
systematic=syst)
elif var == 'e0eta':
if ch in [
'mmSSonZ', 'mmSSoffZ', 'mmmSSoffZ',
'mmmSSonZ', 'mmmm'
]:
continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut,
e0eta=values,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
weight=weights_flat,
systematic=syst)
elif var == 'm0eta':
if ch in [
'eeSSonZ', 'eeSSoffZ', 'eeeSSoffZ',
'eeeSSonZ', 'eeee'
]:
continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut,
m0eta=values,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
weight=weights_flat,
systematic=syst)
elif var == 'j0pt':
if lev == 'base': continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut,
j0pt=values,
sample=histAxisName,
channel=ch,
cut=lev,
sumcharge=sumcharge,
weight=weights_flat,
systematic=syst)
return hout
def process(self, events):
output = self.accumulator.identity()
# we can use a very loose preselection to filter the events. nothing is done with this presel, though
presel = ak.num(events.Jet) > 0
if self.year == 2016:
lumimask = LumiMask(
'../data/lumi/Cert_271036-284044_13TeV_Legacy2016_Collisions16_JSON.txt'
)
if self.year == 2017:
lumimask = LumiMask(
'../data/lumi/Cert_294927-306462_13TeV_UL2017_Collisions17_GoldenJSON.txt'
)
if self.year == 2018:
lumimask = LumiMask(
'../data/lumi/Cert_314472-325175_13TeV_Legacy2018_Collisions18_JSON.txt'
)
ev = events[presel]
dataset = ev.metadata['dataset']
# load the config - probably not needed anymore
cfg = loadConfig()
output['totalEvents']['all'] += len(events)
output['skimmedEvents']['all'] += len(ev)
if self.year == 2018:
triggers = ev.HLT.Ele23_Ele12_CaloIdL_TrackIdL_IsoVL
elif self.year == 2017:
triggers = ev.HLT.Ele23_Ele12_CaloIdL_TrackIdL_IsoVL
elif self.year == 2016:
triggers = ev.HLT.Ele23_Ele12_CaloIdL_TrackIdL_IsoVL_DZ
## Electrons
electron = Collections(ev, "Electron", "tightFCNC", 0, self.year).get()
electron = electron[(electron.pt > 25) & (np.abs(electron.eta) < 2.4)]
loose_electron = Collections(ev, "Electron", "looseFCNC", 0,
self.year).get()
loose_electron = loose_electron[(loose_electron.pt > 25)
& (np.abs(loose_electron.eta) < 2.4)]
SSelectron = (ak.sum(electron.charge, axis=1) != 0) & (ak.num(electron)
== 2)
OSelectron = (ak.sum(electron.charge, axis=1) == 0) & (ak.num(electron)
== 2)
dielectron = choose(electron, 2)
dielectron_mass = (dielectron['0'] + dielectron['1']).mass
dielectron_pt = (dielectron['0'] + dielectron['1']).pt
leading_electron_idx = ak.singletons(ak.argmax(electron.pt, axis=1))
leading_electron = electron[(leading_electron_idx)]
leading_electron = leading_electron[(leading_electron.pt > 30)]
trailing_electron_idx = ak.singletons(ak.argmin(electron.pt, axis=1))
trailing_electron = electron[trailing_electron_idx]
##Muons
loose_muon = Collections(ev, "Muon", "looseFCNC", 0, self.year).get()
loose_muon = loose_muon[(loose_muon.pt > 20)
& (np.abs(loose_muon.eta) < 2.4)]
#jets
jet = getJets(ev, minPt=40, maxEta=2.4, pt_var='pt')
jet = jet[~match(jet, loose_muon,
deltaRCut=0.4)] # remove jets that overlap with muons
jet = jet[~match(
jet, electron,
deltaRCut=0.4)] # remove jets that overlap with electrons
## MET -> can switch to puppi MET
met_pt = ev.MET.pt
met_phi = ev.MET.phi
#weights
weight = Weights(len(ev))
weight2 = Weights(len(ev))
weight2.add("charge flip",
self.charge_flip_ratio.flip_weight(electron))
#selections
filters = getFilters(ev, year=self.year, dataset=dataset, UL=False)
mask = lumimask(ev.run, ev.luminosityBlock)
ss = (SSelectron)
os = (OSelectron)
mass = (ak.min(np.abs(dielectron_mass - 91.2), axis=1) < 15)
lead_electron = (ak.min(leading_electron.pt, axis=1) > 30)
jet1 = (ak.num(jet) >= 1)
jet2 = (ak.num(jet) >= 2)
num_loose = ((ak.num(loose_electron) == 2) & (ak.num(loose_muon) == 0))
selection = PackedSelection()
selection.add('filter', (filters))
selection.add('mask', (mask))
selection.add('ss', ss)
selection.add('os', os)
selection.add('mass', mass)
selection.add('leading', lead_electron)
selection.add('triggers', triggers)
selection.add('one jet', jet1)
selection.add('two jets', jet2)
selection.add('num_loose', num_loose)
bl_reqs = ['filter'] + ['triggers'] + ['mask']
bl_reqs_d = {sel: True for sel in bl_reqs}
baseline = selection.require(**bl_reqs_d)
s_reqs = bl_reqs + ['ss'] + ['mass'] + ['num_loose'] + ['leading']
s_reqs_d = {sel: True for sel in s_reqs}
ss_sel = selection.require(**s_reqs_d)
o_reqs = bl_reqs + ['os'] + ['mass'] + ['num_loose'] + ['leading']
o_reqs_d = {sel: True for sel in o_reqs}
os_sel = selection.require(**o_reqs_d)
j1s_reqs = s_reqs + ['one jet']
j1s_reqs_d = {sel: True for sel in j1s_reqs}
j1ss_sel = selection.require(**j1s_reqs_d)
j1o_reqs = o_reqs + ['one jet']
j1o_reqs_d = {sel: True for sel in j1o_reqs}
j1os_sel = selection.require(**j1o_reqs_d)
j2s_reqs = s_reqs + ['two jets']
j2s_reqs_d = {sel: True for sel in j2s_reqs}
j2ss_sel = selection.require(**j2s_reqs_d)
j2o_reqs = o_reqs + ['two jets']
j2o_reqs_d = {sel: True for sel in j2o_reqs}
j2os_sel = selection.require(**j2o_reqs_d)
#outputs
output["electron_data1"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(leading_electron[os_sel].pt)),
eta=ak.to_numpy(ak.flatten(leading_electron[os_sel].eta)),
phi=ak.to_numpy(ak.flatten(leading_electron[os_sel].phi)),
weight=weight2.weight()[os_sel])
output["electron_data2"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(trailing_electron[os_sel].pt)),
eta=ak.to_numpy(ak.flatten(trailing_electron[os_sel].eta)),
phi=ak.to_numpy(ak.flatten(trailing_electron[os_sel].phi)),
weight=weight2.weight()[os_sel])
output["electron_data3"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(leading_electron[j1os_sel].pt)),
eta=ak.to_numpy(ak.flatten(leading_electron[j1os_sel].eta)),
phi=ak.to_numpy(ak.flatten(leading_electron[j1os_sel].phi)),
weight=weight2.weight()[j1os_sel])
output["electron_data4"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(trailing_electron[j1os_sel].pt)),
eta=ak.to_numpy(ak.flatten(trailing_electron[j1os_sel].eta)),
phi=ak.to_numpy(ak.flatten(trailing_electron[j1os_sel].phi)),
weight=weight2.weight()[j1os_sel])
output["electron_data5"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(leading_electron[j2os_sel].pt)),
eta=ak.to_numpy(ak.flatten(leading_electron[j2os_sel].eta)),
phi=ak.to_numpy(ak.flatten(leading_electron[j2os_sel].phi)),
weight=weight2.weight()[j2os_sel])
output["electron_data6"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(trailing_electron[j2os_sel].pt)),
eta=ak.to_numpy(ak.flatten(trailing_electron[j2os_sel].eta)),
phi=ak.to_numpy(ak.flatten(trailing_electron[j2os_sel].phi)),
weight=weight2.weight()[j2os_sel])
output["electron_data7"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(leading_electron[ss_sel].pt)),
eta=ak.to_numpy(ak.flatten(leading_electron[ss_sel].eta)),
phi=ak.to_numpy(ak.flatten(leading_electron[ss_sel].phi)),
weight=weight.weight()[ss_sel])
output["electron_data8"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(trailing_electron[ss_sel].pt)),
eta=ak.to_numpy(ak.flatten(trailing_electron[ss_sel].eta)),
phi=ak.to_numpy(ak.flatten(trailing_electron[ss_sel].phi)),
weight=weight.weight()[ss_sel])
output["electron_data9"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(leading_electron[j1ss_sel].pt)),
eta=ak.to_numpy(ak.flatten(leading_electron[j1ss_sel].eta)),
phi=ak.to_numpy(ak.flatten(leading_electron[j1ss_sel].phi)),
weight=weight.weight()[j1ss_sel])
output["electron_data10"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(trailing_electron[j1ss_sel].pt)),
eta=ak.to_numpy(ak.flatten(trailing_electron[j1ss_sel].eta)),
phi=ak.to_numpy(ak.flatten(trailing_electron[j1ss_sel].phi)),
weight=weight.weight()[j1ss_sel])
output["electron_data11"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(leading_electron[j2ss_sel].pt)),
eta=ak.to_numpy(ak.flatten(leading_electron[j2ss_sel].eta)),
phi=ak.to_numpy(ak.flatten(leading_electron[j2ss_sel].phi)),
weight=weight.weight()[j2ss_sel])
output["electron_data12"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(trailing_electron[j2ss_sel].pt)),
eta=ak.to_numpy(ak.flatten(trailing_electron[j2ss_sel].eta)),
phi=ak.to_numpy(ak.flatten(trailing_electron[j2ss_sel].phi)),
weight=weight.weight()[j2ss_sel])
output["dilep_mass1"].fill(
dataset=dataset,
mass=ak.to_numpy(ak.flatten(dielectron_mass[os_sel])),
pt=ak.to_numpy(ak.flatten(dielectron_pt[os_sel])),
weight=weight2.weight()[os_sel])
output["dilep_mass2"].fill(
dataset=dataset,
mass=ak.to_numpy(ak.flatten(dielectron_mass[j1os_sel])),
pt=ak.to_numpy(ak.flatten(dielectron_pt[j1os_sel])),
weight=weight2.weight()[j1os_sel])
output["dilep_mass3"].fill(
dataset=dataset,
mass=ak.to_numpy(ak.flatten(dielectron_mass[j2os_sel])),
pt=ak.to_numpy(ak.flatten(dielectron_pt[j2os_sel])),
weight=weight2.weight()[j2os_sel])
output["dilep_mass4"].fill(
dataset=dataset,
mass=ak.to_numpy(ak.flatten(dielectron_mass[ss_sel])),
pt=ak.to_numpy(ak.flatten(dielectron_pt[ss_sel])),
weight=weight.weight()[ss_sel])
output["dilep_mass5"].fill(
dataset=dataset,
mass=ak.to_numpy(ak.flatten(dielectron_mass[j1ss_sel])),
pt=ak.to_numpy(ak.flatten(dielectron_pt[j1ss_sel])),
weight=weight.weight()[j1ss_sel])
output["dilep_mass6"].fill(
dataset=dataset,
mass=ak.to_numpy(ak.flatten(dielectron_mass[j2ss_sel])),
pt=ak.to_numpy(ak.flatten(dielectron_pt[j2ss_sel])),
weight=weight.weight()[j2ss_sel])
output["MET"].fill(dataset=dataset,
pt=met_pt[os_sel],
weight=weight2.weight()[os_sel])
output["MET2"].fill(dataset=dataset,
pt=met_pt[j1os_sel],
weight=weight2.weight()[j1os_sel])
output["MET3"].fill(dataset=dataset,
pt=met_pt[j2os_sel],
weight=weight2.weight()[j2os_sel])
output["MET4"].fill(dataset=dataset,
pt=met_pt[ss_sel],
weight=weight.weight()[ss_sel])
output["MET5"].fill(dataset=dataset,
pt=met_pt[j1ss_sel],
weight=weight.weight()[j1ss_sel])
output["MET6"].fill(dataset=dataset,
pt=met_pt[j2ss_sel],
weight=weight.weight()[j2ss_sel])
output["N_jet"].fill(dataset=dataset,
multiplicity=ak.num(jet)[os_sel],
weight=weight2.weight()[os_sel])
output["N_jet2"].fill(dataset=dataset,
multiplicity=ak.num(jet)[j1os_sel],
weight=weight2.weight()[j1os_sel])
output["N_jet3"].fill(dataset=dataset,
multiplicity=ak.num(jet)[j2os_sel],
weight=weight2.weight()[j2os_sel])
output["N_jet4"].fill(dataset=dataset,
multiplicity=ak.num(jet)[ss_sel],
weight=weight.weight()[ss_sel])
output["N_jet5"].fill(dataset=dataset,
multiplicity=ak.num(jet)[j1ss_sel],
weight=weight.weight()[j1ss_sel])
output["N_jet6"].fill(dataset=dataset,
multiplicity=ak.num(jet)[j2ss_sel],
weight=weight.weight()[j2ss_sel])
output["PV_npvsGood"].fill(dataset=dataset,
multiplicity=ev.PV[os_sel].npvsGood,
weight=weight2.weight()[os_sel])
output["PV_npvsGood2"].fill(dataset=dataset,
multiplicity=ev.PV[j1os_sel].npvsGood,
weight=weight2.weight()[j1os_sel])
output["PV_npvsGood3"].fill(dataset=dataset,
multiplicity=ev.PV[j2os_sel].npvsGood,
weight=weight2.weight()[j2os_sel])
output["PV_npvsGood4"].fill(dataset=dataset,
multiplicity=ev.PV[ss_sel].npvsGood,
weight=weight.weight()[ss_sel])
output["PV_npvsGood5"].fill(dataset=dataset,
multiplicity=ev.PV[j1ss_sel].npvsGood,
weight=weight.weight()[j1ss_sel])
output["PV_npvsGood6"].fill(dataset=dataset,
multiplicity=ev.PV[j2ss_sel].npvsGood,
weight=weight.weight()[j2ss_sel])
return output
def process(self, events):
# Dataset parameters
dataset = events.metadata['dataset']
year = self._samples[dataset]['year']
xsec = self._samples[dataset]['xsec']
sow = self._samples[dataset]['nSumOfWeights']
isData = self._samples[dataset]['isData']
datasets = [
'SingleMuon', 'SingleElectron', 'EGamma', 'MuonEG', 'DoubleMuon',
'DoubleElectron'
]
for d in datasets:
if d in dataset: dataset = dataset.split('_')[0]
# Initialize objects
met = events.MET
e = events.Electron
mu = events.Muon
tau = events.Tau
j = events.Jet
# Muon selection
#mu['isGood'] = isMuonMVA(mu.pt, mu.eta, mu.dxy, mu.dz, mu.miniPFRelIso_all, mu.sip3d, mu.mvaTTH, mu.mediumPromptId, mu.tightCharge, minpt=10)
mu['isPres'] = isPresMuon(mu.dxy, mu.dz, mu.sip3d, mu.looseId)
mu['isTight'] = isTightMuon(mu.pt,
mu.eta,
mu.dxy,
mu.dz,
mu.pfRelIso03_all,
mu.sip3d,
mu.mvaTTH,
mu.mediumPromptId,
mu.tightCharge,
mu.looseId,
minpt=10)
mu['isGood'] = mu['isPres'] & mu['isTight']
leading_mu = mu[ak.argmax(mu.pt, axis=-1, keepdims=True)]
leading_mu = leading_mu[leading_mu.isGood]
mu = mu[mu.isGood]
mu_pres = mu[mu.isPres]
# Electron selection
#e['isGood'] = isElecMVA(e.pt, e.eta, e.dxy, e.dz, e.miniPFRelIso_all, e.sip3d, e.mvaTTH, e.mvaFall17V2Iso, e.lostHits, e.convVeto, e.tightCharge, minpt=10)
e['isPres'] = isPresElec(e.pt,
e.eta,
e.dxy,
e.dz,
e.miniPFRelIso_all,
e.sip3d,
e.lostHits,
minpt=15)
e['isTight'] = isTightElec(e.pt,
e.eta,
e.dxy,
e.dz,
e.miniPFRelIso_all,
e.sip3d,
e.mvaTTH,
e.mvaFall17V2Iso,
e.lostHits,
e.convVeto,
e.tightCharge,
e.sieie,
e.hoe,
e.eInvMinusPInv,
minpt=15)
e['isClean'] = isClean(e, mu, drmin=0.05)
e['isGood'] = e['isPres'] & e['isTight'] & e['isClean']
leading_e = e[ak.argmax(e.pt, axis=-1, keepdims=True)]
leading_e = leading_e[leading_e.isGood]
e = e[e.isGood]
e_pres = e[e.isPres & e.isClean]
# Tau selection
tau['isPres'] = isPresTau(tau.pt,
tau.eta,
tau.dxy,
tau.dz,
tau.leadTkPtOverTauPt,
tau.idAntiMu,
tau.idAntiEle,
tau.rawIso,
tau.idDecayModeNewDMs,
minpt=20)
tau['isClean'] = isClean(tau, e_pres, drmin=0.4) & isClean(
tau, mu_pres, drmin=0.4)
tau['isGood'] = tau['isPres'] # & tau['isClean'], for the moment
tau = tau[tau.isGood]
nElec = ak.num(e)
nMuon = ak.num(mu)
nTau = ak.num(tau)
twoLeps = (nElec + nMuon) == 2
threeLeps = (nElec + nMuon) == 3
twoElec = (nElec == 2)
twoMuon = (nMuon == 2)
e0 = e[ak.argmax(e.pt, axis=-1, keepdims=True)]
m0 = mu[ak.argmax(mu.pt, axis=-1, keepdims=True)]
# Jet selection
jetptname = 'pt_nom' if hasattr(j, 'pt_nom') else 'pt'
j['isGood'] = isTightJet(getattr(j,
jetptname), j.eta, j.jetId, j.neHEF,
j.neEmEF, j.chHEF, j.chEmEF, j.nConstituents)
#j['isgood'] = isGoodJet(j.pt, j.eta, j.jetId)
#j['isclean'] = isClean(j, e, mu)
j['isClean'] = isClean(j, e, drmin=0.4) & isClean(
j, mu, drmin=0.4) # & isClean(j, tau, drmin=0.4)
goodJets = j[(j.isClean) & (j.isGood)]
njets = ak.num(goodJets)
ht = ak.sum(goodJets.pt, axis=-1)
j0 = goodJets[ak.argmax(goodJets.pt, axis=-1, keepdims=True)]
#nbtags = ak.num(goodJets[goodJets.btagDeepFlavB > 0.2770])
nbtags = ak.num(goodJets[goodJets.btagDeepB > 0.4941])
##################################################################
### 2 same-sign leptons
##################################################################
# emu
singe = e[(nElec == 1) & (nMuon == 1) & (e.pt > -1)]
singm = mu[(nElec == 1) & (nMuon == 1) & (mu.pt > -1)]
em = ak.cartesian({"e": singe, "m": singm})
emSSmask = (em.e.charge * em.m.charge > 0)
emSS = em[emSSmask]
nemSS = len(ak.flatten(emSS))
# ee and mumu
# pt>-1 to preserve jagged dimensions
ee = e[(nElec == 2) & (nMuon == 0) & (e.pt > -1)]
mm = mu[(nElec == 0) & (nMuon == 2) & (mu.pt > -1)]
eepairs = ak.combinations(ee, 2, fields=["e0", "e1"])
eeSSmask = (eepairs.e0.charge * eepairs.e1.charge > 0)
eeonZmask = (np.abs((eepairs.e0 + eepairs.e1).mass - 91.2) < 10)
eeoffZmask = (eeonZmask == 0)
mmpairs = ak.combinations(mm, 2, fields=["m0", "m1"])
mmSSmask = (mmpairs.m0.charge * mmpairs.m1.charge > 0)
mmonZmask = (np.abs((mmpairs.m0 + mmpairs.m1).mass - 91.2) < 10)
mmoffZmask = (mmonZmask == 0)
eeSSonZ = eepairs[eeSSmask & eeonZmask]
eeSSoffZ = eepairs[eeSSmask & eeoffZmask]
mmSSonZ = mmpairs[mmSSmask & mmonZmask]
mmSSoffZ = mmpairs[mmSSmask & mmoffZmask]
neeSS = len(ak.flatten(eeSSonZ)) + len(ak.flatten(eeSSoffZ))
nmmSS = len(ak.flatten(mmSSonZ)) + len(ak.flatten(mmSSoffZ))
print('Same-sign events [ee, emu, mumu] = [%i, %i, %i]' %
(neeSS, nemSS, nmmSS))
# Cuts
eeSSmask = (ak.num(eeSSmask[eeSSmask]) > 0)
mmSSmask = (ak.num(mmSSmask[mmSSmask]) > 0)
eeonZmask = (ak.num(eeonZmask[eeonZmask]) > 0)
eeoffZmask = (ak.num(eeoffZmask[eeoffZmask]) > 0)
mmonZmask = (ak.num(mmonZmask[mmonZmask]) > 0)
mmoffZmask = (ak.num(mmoffZmask[mmoffZmask]) > 0)
emSSmask = (ak.num(emSSmask[emSSmask]) > 0)
##################################################################
### 3 leptons
##################################################################
# eem
muon_eem = mu[(nElec == 2) & (nMuon == 1) & (mu.pt > -1)]
elec_eem = e[(nElec == 2) & (nMuon == 1) & (e.pt > -1)]
ee_eem = ak.combinations(elec_eem, 2, fields=["e0", "e1"])
ee_eemZmask = (ee_eem.e0.charge * ee_eem.e1.charge < 1) & (np.abs(
(ee_eem.e0 + ee_eem.e1).mass - 91.2) < 10)
ee_eemOffZmask = (ee_eem.e0.charge * ee_eem.e1.charge < 1) & (np.abs(
(ee_eem.e0 + ee_eem.e1).mass - 91.2) > 10)
ee_eemZmask = (ak.num(ee_eemZmask[ee_eemZmask]) > 0)
ee_eemOffZmask = (ak.num(ee_eemOffZmask[ee_eemOffZmask]) > 0)
eepair_eem = (ee_eem.e0 + ee_eem.e1)
trilep_eem = eepair_eem + muon_eem #ak.cartesian({"e0":ee_eem.e0,"e1":ee_eem.e1, "m":muon_eem})
# mme
muon_mme = mu[(nElec == 1) & (nMuon == 2) & (mu.pt > -1)]
elec_mme = e[(nElec == 1) & (nMuon == 2) & (e.pt > -1)]
mm_mme = ak.combinations(muon_mme, 2, fields=["m0", "m1"])
mm_mmeZmask = (mm_mme.m0.charge * mm_mme.m1.charge < 1) & (np.abs(
(mm_mme.m0 + mm_mme.m1).mass - 91.2) < 10)
mm_mmeOffZmask = (mm_mme.m0.charge * mm_mme.m1.charge < 1) & (np.abs(
(mm_mme.m0 + mm_mme.m1).mass - 91.2) > 10)
mm_mmeZmask = (ak.num(mm_mmeZmask[mm_mmeZmask]) > 0)
mm_mmeOffZmask = (ak.num(mm_mmeOffZmask[mm_mmeOffZmask]) > 0)
mmpair_mme = (mm_mme.m0 + mm_mme.m1)
trilep_mme = mmpair_mme + elec_mme
mZ_mme = mmpair_mme.mass
mZ_eem = eepair_eem.mass
m3l_eem = trilep_eem.mass
m3l_mme = trilep_mme.mass
# eee and mmm
eee = e[(nElec == 3) & (nMuon == 0) & (e.pt > -1)]
mmm = mu[(nElec == 0) & (nMuon == 3) & (mu.pt > -1)]
eee_leps = ak.combinations(eee, 3, fields=["e0", "e1", "e2"])
mmm_leps = ak.combinations(mmm, 3, fields=["m0", "m1", "m2"])
ee_pairs = ak.combinations(eee, 2, fields=["e0", "e1"])
mm_pairs = ak.combinations(mmm, 2, fields=["m0", "m1"])
ee_pairs_index = ak.argcombinations(eee, 2, fields=["e0", "e1"])
mm_pairs_index = ak.argcombinations(mmm, 2, fields=["m0", "m1"])
mmSFOS_pairs = mm_pairs[
(np.abs(mm_pairs.m0.pdgId) == np.abs(mm_pairs.m1.pdgId))
& (mm_pairs.m0.charge != mm_pairs.m1.charge)]
offZmask_mm = ak.all(
np.abs((mmSFOS_pairs.m0 + mmSFOS_pairs.m1).mass - 91.2) > 10.,
axis=1,
keepdims=True) & (ak.num(mmSFOS_pairs) > 0)
onZmask_mm = ak.any(
np.abs((mmSFOS_pairs.m0 + mmSFOS_pairs.m1).mass - 91.2) < 10.,
axis=1,
keepdims=True)
eeSFOS_pairs = ee_pairs[
(np.abs(ee_pairs.e0.pdgId) == np.abs(ee_pairs.e1.pdgId))
& (ee_pairs.e0.charge != ee_pairs.e1.charge)]
offZmask_ee = ak.all(
np.abs((eeSFOS_pairs.e0 + eeSFOS_pairs.e1).mass - 91.2) > 10,
axis=1,
keepdims=True) & (ak.num(eeSFOS_pairs) > 0)
onZmask_ee = ak.any(
np.abs((eeSFOS_pairs.e0 + eeSFOS_pairs.e1).mass - 91.2) < 10,
axis=1,
keepdims=True)
# Create masks **for event selection**
eeeOnZmask = (ak.num(onZmask_ee[onZmask_ee]) > 0)
eeeOffZmask = (ak.num(offZmask_ee[offZmask_ee]) > 0)
mmmOnZmask = (ak.num(onZmask_mm[onZmask_mm]) > 0)
mmmOffZmask = (ak.num(offZmask_mm[offZmask_mm]) > 0)
# Now we need to create masks for the leptons in order to select leptons from the Z boson candidate (in onZ categories)
ZeeMask = ak.argmin(np.abs((eeSFOS_pairs.e0 + eeSFOS_pairs.e1).mass -
91.2),
axis=1,
keepdims=True)
ZmmMask = ak.argmin(np.abs((mmSFOS_pairs.m0 + mmSFOS_pairs.m1).mass -
91.2),
axis=1,
keepdims=True)
Zee = eeSFOS_pairs[ZeeMask]
Zmm = mmSFOS_pairs[ZmmMask]
eZ0 = Zee.e0[ak.num(eeSFOS_pairs) > 0]
eZ1 = Zee.e1[ak.num(eeSFOS_pairs) > 0]
eZ = eZ0 + eZ1
mZ0 = Zmm.m0[ak.num(mmSFOS_pairs) > 0]
mZ1 = Zmm.m1[ak.num(mmSFOS_pairs) > 0]
mZ = mZ0 + mZ1
mZ_eee = eZ.mass
mZ_mmm = mZ.mass
# And for the W boson
ZmmIndices = mm_pairs_index[ZmmMask]
ZeeIndices = ee_pairs_index[ZeeMask]
eW = eee[~ZeeIndices.e0 | ~ZeeIndices.e1]
mW = mmm[~ZmmIndices.m0 | ~ZmmIndices.m1]
triElec = eee_leps.e0 + eee_leps.e1 + eee_leps.e2
triMuon = mmm_leps.m0 + mmm_leps.m1 + mmm_leps.m2
m3l_eee = triElec.mass
m3l_mmm = triMuon.mass
# Triggers
trig_eeSS = passTrigger(events, 'ee', isData, dataset)
trig_mmSS = passTrigger(events, 'mm', isData, dataset)
trig_emSS = passTrigger(events, 'em', isData, dataset)
trig_eee = passTrigger(events, 'eee', isData, dataset)
trig_mmm = passTrigger(events, 'mmm', isData, dataset)
trig_eem = passTrigger(events, 'eem', isData, dataset)
trig_mme = passTrigger(events, 'mme', isData, dataset)
# MET filters
# Weights
genw = np.ones_like(
events['MET_pt']) if isData else events['genWeight']
weights = coffea.analysis_tools.Weights(len(events))
weights.add('norm', genw if isData else (xsec / sow) * genw)
eftweights = events['EFTfitCoefficients'] if hasattr(
events, "EFTfitCoefficients") else []
# Selections and cuts
selections = PackedSelection()
channels2LSS = ['eeSSonZ', 'eeSSoffZ', 'mmSSonZ', 'mmSSoffZ', 'emSS']
selections.add('eeSSonZ', (eeonZmask) & (eeSSmask) & (trig_eeSS))
selections.add('eeSSoffZ', (eeoffZmask) & (eeSSmask) & (trig_eeSS))
selections.add('mmSSonZ', (mmonZmask) & (mmSSmask) & (trig_mmSS))
selections.add('mmSSoffZ', (mmoffZmask) & (mmSSmask) & (trig_mmSS))
selections.add('emSS', (emSSmask) & (trig_emSS))
channels3L = ['eemSSonZ', 'eemSSoffZ', 'mmeSSonZ', 'mmeSSoffZ']
selections.add('eemSSonZ', (ee_eemZmask) & (trig_eem))
selections.add('eemSSoffZ', (ee_eemOffZmask) & (trig_eem))
selections.add('mmeSSonZ', (mm_mmeZmask) & (trig_mme))
selections.add('mmeSSoffZ', (mm_mmeOffZmask) & (trig_mme))
channels3L += ['eeeSSonZ', 'eeeSSoffZ', 'mmmSSonZ', 'mmmSSoffZ']
selections.add('eeeSSonZ', (eeeOnZmask) & (trig_eee))
selections.add('eeeSSoffZ', (eeeOffZmask) & (trig_eee))
selections.add('mmmSSonZ', (mmmOnZmask) & (trig_mmm))
selections.add('mmmSSoffZ', (mmmOffZmask) & (trig_mmm))
levels = ['base', '2jets', '4jets', '4j1b', '4j2b']
selections.add('base', (nElec + nMuon >= 2))
selections.add('2jets', (njets >= 2))
selections.add('4jets', (njets >= 4))
selections.add('4j1b', (njets >= 4) & (nbtags >= 1))
selections.add('4j2b', (njets >= 4) & (nbtags >= 2))
# Variables
invMass_eeSSonZ = (eeSSonZ.e0 + eeSSonZ.e1).mass
invMass_eeSSoffZ = (eeSSoffZ.e0 + eeSSoffZ.e1).mass
invMass_mmSSonZ = (mmSSonZ.m0 + mmSSonZ.m1).mass
invMass_mmSSoffZ = (mmSSoffZ.m0 + mmSSoffZ.m1).mass
invMass_emSS = (emSS.e + emSS.m).mass
varnames = {}
varnames['met'] = met.pt
varnames['ht'] = ht
varnames['njets'] = njets
varnames['nbtags'] = nbtags
varnames['invmass'] = {
'eeSSonZ': invMass_eeSSonZ,
'eeSSoffZ': invMass_eeSSoffZ,
'mmSSonZ': invMass_mmSSonZ,
'mmSSoffZ': invMass_mmSSoffZ,
'emSS': invMass_emSS,
'eemSSonZ': mZ_eem,
'eemSSoffZ': mZ_eem,
'mmeSSonZ': mZ_mme,
'mmeSSoffZ': mZ_mme,
'eeeSSonZ': mZ_eee,
'eeeSSoffZ': mZ_eee,
'mmmSSonZ': mZ_mmm,
'mmmSSoffZ': mZ_mmm,
}
varnames['m3l'] = {
'eemSSonZ': m3l_eem,
'eemSSoffZ': m3l_eem,
'mmeSSonZ': m3l_mme,
'mmeSSoffZ': m3l_mme,
'eeeSSonZ': m3l_eee,
'eeeSSoffZ': m3l_eee,
'mmmSSonZ': m3l_mmm,
'mmmSSoffZ': m3l_mmm,
}
varnames['e0pt'] = e0.pt
varnames['e0eta'] = e0.eta
varnames['m0pt'] = m0.pt
varnames['m0eta'] = m0.eta
varnames['j0pt'] = j0.pt
varnames['j0eta'] = j0.eta
varnames['counts'] = np.ones_like(events.MET.pt)
# fill Histos
hout = self.accumulator.identity()
allweights = weights.weight().flatten(
) # Why does it not complain about .flatten() here?
hout['SumOfEFTweights'].fill(eftweights,
sample=dataset,
SumOfEFTweights=varnames['counts'],
weight=allweights)
for var, v in varnames.items():
for ch in channels2LSS + channels3L:
for lev in levels:
weight = weights.weight()
cuts = [ch] + [lev]
cut = selections.all(*cuts)
weights_flat = weight[cut].flatten(
) # Why does it not complain about .flatten() here?
weights_ones = np.ones_like(weights_flat, dtype=np.int)
eftweightsvalues = eftweights[cut] if len(
eftweights) > 0 else []
if var == 'invmass':
if ch in ['eeeSSoffZ', 'mmmSSoffZ']: continue
elif ch in ['eeeSSonZ', 'mmmSSonZ']:
continue #values = v[ch]
else:
values = ak.flatten(v[ch][cut])
hout['invmass'].fill(sample=dataset,
channel=ch,
cut=lev,
invmass=values,
weight=weights_flat)
elif var == 'm3l':
if ch in [
'eeSSonZ', 'eeSSoffZ', 'mmSSonZ', 'mmSSoffZ',
'emSS', 'eeeSSoffZ', 'mmmSSoffZ', 'eeeSSonZ',
'mmmSSonZ'
]:
continue
values = ak.flatten(v[ch][cut])
hout['m3l'].fill(eftweightsvalues,
sample=dataset,
channel=ch,
cut=lev,
m3l=values,
weight=weights_flat)
else:
values = v[cut]
if var == 'ht':
hout[var].fill(eftweightsvalues,
ht=values,
sample=dataset,
channel=ch,
cut=lev,
weight=weights_flat)
elif var == 'met':
hout[var].fill(eftweightsvalues,
met=values,
sample=dataset,
channel=ch,
cut=lev,
weight=weights_flat)
elif var == 'njets':
hout[var].fill(eftweightsvalues,
njets=values,
sample=dataset,
channel=ch,
cut=lev,
weight=weights_flat)
elif var == 'nbtags':
hout[var].fill(eftweightsvalues,
nbtags=values,
sample=dataset,
channel=ch,
cut=lev,
weight=weights_flat)
elif var == 'counts':
hout[var].fill(counts=values,
sample=dataset,
channel=ch,
cut=lev,
weight=weights_ones)
elif var == 'j0eta':
if lev == 'base': continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(eftweightsvalues,
j0eta=values,
sample=dataset,
channel=ch,
cut=lev,
weight=weights_flat)
elif var == 'e0pt':
if ch in [
'mmSSonZ', 'mmSSoffZ', 'mmmSSoffZ',
'mmmSSonZ'
]:
continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(
eftweightsvalues,
e0pt=values,
sample=dataset,
channel=ch,
cut=lev,
weight=weights_flat
) # Crashing here, not sure why. Related to values?
elif var == 'm0pt':
if ch in [
'eeSSonZ', 'eeSSoffZ', 'eeeSSoffZ',
'eeeSSonZ'
]:
continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(eftweightsvalues,
m0pt=values,
sample=dataset,
channel=ch,
cut=lev,
weight=weights_flat)
elif var == 'e0eta':
if ch in [
'mmSSonZ', 'mmSSoffZ', 'mmmSSoffZ',
'mmmSSonZ'
]:
continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(eftweightsvalues,
e0eta=values,
sample=dataset,
channel=ch,
cut=lev,
weight=weights_flat)
elif var == 'm0eta':
if ch in [
'eeSSonZ', 'eeSSoffZ', 'eeeSSoffZ',
'eeeSSonZ'
]:
continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(eftweightsvalues,
m0eta=values,
sample=dataset,
channel=ch,
cut=lev,
weight=weights_flat)
elif var == 'j0pt':
if lev == 'base': continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(eftweightsvalues,
j0pt=values,
sample=dataset,
channel=ch,
cut=lev,
weight=weights_flat)
return hout
def process(self, events):
# Dataset parameters
dataset = events.metadata['dataset']
year = self._samples[dataset]['year']
xsec = self._samples[dataset]['xsec']
sow = self._samples[dataset]['nSumOfWeights']
isData = self._samples[dataset]['isData']
datasets = [
'SingleMuon', 'SingleElectron', 'EGamma', 'MuonEG', 'DoubleMuon',
'DoubleElectron'
]
for d in datasets:
if d in dataset: dataset = dataset.split('_')[0]
# Extract the EFT quadratic coefficients and optionally use them to calculate the coefficients on the w**2 quartic function
# eft_coeffs is never Jagged so convert immediately to numpy for ease of use.
eft_coeffs = ak.to_numpy(events['EFTfitCoefficients']) if hasattr(
events, "EFTfitCoefficients") else None
if eft_coeffs is not None:
# Check to see if the ordering of WCs for this sample matches what want
if self._samples[dataset]['WCnames'] != self._wc_names_lst:
eft_coeffs = efth.remap_coeffs(
self._samples[dataset]['WCnames'], self._wc_names_lst,
eft_coeffs)
eft_w2_coeffs = efth.calc_w2_coeffs(eft_coeffs, self._dtype) if (
self._do_errors and eft_coeffs is not None) else None
# Initialize objects (GEN objects)
e = events.GenPart[abs(events.GenPart.pdgId) == 11]
m = events.GenPart[abs(events.GenPart.pdgId) == 13]
tau = events.GenPart[abs(events.GenPart.pdgId) == 15]
j = events.GenJet
run = events.run
luminosityBlock = events.luminosityBlock
event = events.event
print("\n\nInfo about events:")
print("\trun:", run)
print("\tluminosityBlock:", luminosityBlock)
print("\tevent:", event)
print("\nLeptons before selection:")
print("\te pt", e.pt)
print("\te eta", e.eta)
print("\tm pt", m.pt)
print("\tm eta", m.eta)
######## Lep selection ########
e_selec = ((e.pt > 15) & (abs(e.eta) < 2.5))
m_selec = ((m.pt > 15) & (abs(m.eta) < 2.5))
e = e[e_selec]
m = m[m_selec]
# Put the e and mu togheter
l = ak.concatenate([e, m], axis=1)
n_e = ak.num(e)
n_m = ak.num(m)
n_l = ak.num(l)
at_least_two_leps = (n_l >= 2)
e0 = e[ak.argmax(e.pt, axis=-1, keepdims=True)]
m0 = m[ak.argmax(m.pt, axis=-1, keepdims=True)]
l0 = l[ak.argmax(l.pt, axis=-1, keepdims=True)]
print("\nLeptons after selection:")
print("\te pt", e.pt)
print("\tm pt", m.pt)
print("\tl pt:", l.pt)
print("\tn e", n_e)
print("\tn m", n_m)
print("\tn l", n_l)
print("\nMask for at least two lep:", at_least_two_leps)
print("\nLeading lepton info:")
print("\te0", e0.pt)
print("\tm0", m0.pt)
print("\tl0", l0.pt)
######## Jet selection ########
print("\nJet info:")
print("\tjpt before selection", j.pt)
j_selec = ((j.pt > 30) & (abs(j.eta) < 2.5))
print("\tjselect", j_selec)
j = j[j_selec]
print("\tjpt", j.pt)
j['isClean'] = isClean(j, e, drmin=0.4) & isClean(j, m, drmin=0.4)
j_isclean = isClean(j, e, drmin=0.4) & isClean(j, m, drmin=0.4)
print("\tj is clean", j_isclean)
j = j[j_isclean]
print("\tclean jets pt", j.pt)
n_j = ak.num(j)
print("\tn_j", n_j)
j0 = j[ak.argmax(j.pt, axis=-1, keepdims=True)]
print("\tj0pt", j0.pt)
at_least_two_jets = (n_j >= 2)
print("\tat_least_two_jets", at_least_two_jets)
######## Selections and cuts ########
event_selec = (at_least_two_leps & at_least_two_jets)
print("\nEvent selection:", event_selec, "\n")
selections = PackedSelection()
selections.add('2l2j', event_selec)
varnames = {}
varnames['counts'] = np.ones_like(events.MET.pt)
varnames['njets'] = n_j
varnames['j0pt'] = j0.pt
varnames['j0eta'] = j0.eta
varnames['l0pt'] = l0.pt
######## Fill histos ########
print("\nFilling hists now...\n")
hout = self.accumulator.identity()
for var, v in varnames.items():
cut = selections.all("2l2j")
values = v[cut]
eft_coeffs_cut = eft_coeffs[cut] if eft_coeffs is not None else None
eft_w2_coeffs_cut = eft_w2_coeffs[
cut] if eft_w2_coeffs is not None else None
if var == "counts":
hout[var].fill(counts=values,
sample=dataset,
channel="2l",
cut="2l")
elif var == "njets":
hout[var].fill(njets=values,
sample=dataset,
channel="2l",
cut="2l",
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut)
elif var == "j0pt":
hout[var].fill(j0pt=values,
sample=dataset,
channel="2l",
cut="2l",
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut)
elif var == "j0eta":
hout[var].fill(j0eta=values,
sample=dataset,
channel="2l",
cut="2l",
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut)
elif var == "l0pt":
hout[var].fill(l0pt=values,
sample=dataset,
channel="2l",
cut="2l",
eft_coeff=eft_coeffs_cut,
eft_err_coeff=eft_w2_coeffs_cut)
return hout
def process(self, events):
output = self.accumulator.identity()
output['total']['all'] += len(events)
# use a very loose preselection to filter the events
presel = ak.num(events.Jet) > 2
ev = events[presel]
dataset = ev.metadata['dataset']
# load the config - probably not needed anymore
cfg = loadConfig()
## Muons
muon = Collections(ev, "Muon", "vetoTTH").get()
tightmuon = Collections(ev, "Muon", "tightTTH").get()
dimuon = choose(muon, 2)
SSmuon = ak.any((dimuon['0'].charge * dimuon['1'].charge) > 0, axis=1)
leading_muon_idx = ak.singletons(ak.argmax(muon.pt, axis=1))
leading_muon = muon[leading_muon_idx]
## Electrons
electron = Collections(ev, "Electron", "vetoTTH").get()
tightelectron = Collections(ev, "Electron", "tightTTH").get()
dielectron = choose(electron, 2)
SSelectron = ak.any(
(dielectron['0'].charge * dielectron['1'].charge) > 0, axis=1)
leading_electron_idx = ak.singletons(ak.argmax(electron.pt, axis=1))
leading_electron = electron[leading_electron_idx]
## Merge electrons and muons - this should work better now in ak1
dilepton = cross(muon, electron)
SSlepton = ak.any((dilepton['0'].charge * dilepton['1'].charge) > 0,
axis=1)
lepton = ak.concatenate([muon, electron], axis=1)
leading_lepton_idx = ak.singletons(ak.argmax(lepton.pt, axis=1))
leading_lepton = lepton[leading_lepton_idx]
trailing_lepton_idx = ak.singletons(ak.argmin(lepton.pt, axis=1))
trailing_lepton = lepton[trailing_lepton_idx]
dilepton_mass = (leading_lepton + trailing_lepton).mass
dilepton_pt = (leading_lepton + trailing_lepton).pt
dilepton_dR = delta_r(leading_lepton, trailing_lepton)
mt_lep_met = mt(lepton.pt, lepton.phi, ev.MET.pt, ev.MET.phi)
min_mt_lep_met = ak.min(mt_lep_met, axis=1)
## Jets
jet = getJets(ev, minPt=25, maxEta=4.7, pt_var='pt_nom')
jet = jet[ak.argsort(
jet.pt_nom, ascending=False
)] # need to sort wrt smeared and recorrected jet pt
jet = jet[~match(jet, muon,
deltaRCut=0.4)] # remove jets that overlap with muons
jet = jet[~match(
jet, electron,
deltaRCut=0.4)] # remove jets that overlap with electrons
central = jet[(abs(jet.eta) < 2.4)]
btag = getBTagsDeepFlavB(
jet, year=self.year) # should study working point for DeepJet
light = getBTagsDeepFlavB(jet, year=self.year, invert=True)
fwd = getFwdJet(light)
fwd_noPU = getFwdJet(light, puId=False)
tau = getTaus(ev)
track = getIsoTracks(ev)
## forward jets
j_fwd = fwd[ak.singletons(ak.argmax(
fwd.p, axis=1))] # highest momentum spectator
high_score_btag = central[ak.argsort(central.btagDeepFlavB)][:, :2]
bl = cross(lepton, high_score_btag)
bl_dR = delta_r(bl['0'], bl['1'])
min_bl_dR = ak.min(bl_dR, axis=1)
jf = cross(j_fwd, jet)
mjf = (jf['0'] + jf['1']).mass
j_fwd2 = jf[ak.singletons(
ak.argmax(mjf, axis=1)
)]['1'] # this is the jet that forms the largest invariant mass with j_fwd
delta_eta = ak.fill_none(
ak.pad_none(abs(j_fwd2.eta - j_fwd.eta), 1, clip=True), 0)
## MET -> can switch to puppi MET
met_pt = ev.MET.pt
met_phi = ev.MET.phi
## other variables
ht = ak.sum(jet.pt, axis=1)
st = met_pt + ht + ak.sum(muon.pt, axis=1) + ak.sum(electron.pt,
axis=1)
## event selectors
filters = getFilters(ev, year=self.year, dataset=dataset)
dilep = ((ak.num(tightelectron) + ak.num(tightmuon)) == 2)
lep0pt = ((ak.num(electron[(electron.pt > 25)]) +
ak.num(muon[(muon.pt > 25)])) > 0)
lep1pt = ((ak.num(electron[(electron.pt > 20)]) +
ak.num(muon[(muon.pt > 20)])) > 1)
lepveto = ((ak.num(electron) + ak.num(muon)) == 2)
selection = PackedSelection()
selection.add('lepveto', lepveto)
selection.add('dilep', dilep)
selection.add('filter', (filters))
selection.add('p_T(lep0)>25', lep0pt)
selection.add('p_T(lep1)>20', lep1pt)
selection.add('SS', (SSlepton | SSelectron | SSmuon))
selection.add('N_jet>3', (ak.num(jet) >= 4))
selection.add('N_central>2', (ak.num(central) >= 3))
selection.add('N_btag>0', (ak.num(btag) >= 1))
selection.add('N_fwd>0', (ak.num(fwd) >= 1))
#ss_reqs = ['lepveto', 'dilep', 'filter', 'p_T(lep0)>25', 'p_T(lep1)>20', 'SS']
ss_reqs = [
'lepveto', 'dilep', 'filter', 'p_T(lep0)>25', 'p_T(lep1)>20', 'SS'
]
#bl_reqs = ss_reqs + ['N_jet>3', 'N_central>2', 'N_btag>0', 'N_fwd>0']
bl_reqs = ss_reqs + ['N_jet>3', 'N_central>2', 'N_btag>0']
ss_reqs_d = {sel: True for sel in ss_reqs}
ss_selection = selection.require(**ss_reqs_d)
bl_reqs_d = {sel: True for sel in bl_reqs}
BL = selection.require(**bl_reqs_d)
weight = Weights(len(ev))
if not dataset == 'MuonEG':
# lumi weight
weight.add("weight", ev.weight)
# PU weight - not in the babies...
weight.add("PU",
ev.puWeight,
weightUp=ev.puWeightUp,
weightDown=ev.puWeightDown,
shift=False)
# b-tag SFs
weight.add("btag", self.btagSF.Method1a(btag, light))
# lepton SFs
weight.add("lepton", self.leptonSF.get(electron, muon))
#cutflow = Cutflow(output, ev, weight=weight)
#cutflow_reqs_d = {}
#for req in bl_reqs:
# cutflow_reqs_d.update({req: True})
# cutflow.addRow( req, selection.require(**cutflow_reqs_d) )
labels = {
'topW_v3': 0,
'TTW': 1,
'TTZ': 2,
'TTH': 3,
'ttbar': 4,
'ttbar1l_MG': 4
}
if dataset in labels:
label_mult = labels[dataset]
else:
label_mult = 5
label = np.ones(len(ev[BL])) * label_mult
output["n_lep"] += processor.column_accumulator(
ak.to_numpy((ak.num(electron) + ak.num(muon))[BL]))
output["n_lep_tight"] += processor.column_accumulator(
ak.to_numpy((ak.num(tightelectron) + ak.num(tightmuon))[BL]))
output["lead_lep_pt"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(leading_lepton[BL].pt, axis=1)))
output["lead_lep_eta"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(leading_lepton[BL].eta, axis=1)))
output["lead_lep_phi"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(leading_lepton[BL].phi, axis=1)))
output["lead_lep_charge"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(leading_lepton[BL].charge, axis=1)))
output["sublead_lep_pt"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(trailing_lepton[BL].pt, axis=1)))
output["sublead_lep_eta"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(trailing_lepton[BL].eta, axis=1)))
output["sublead_lep_phi"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(trailing_lepton[BL].phi, axis=1)))
output["sublead_lep_charge"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(trailing_lepton[BL].charge, axis=1)))
output["lead_jet_pt"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(jet[:, 0:1][BL].pt, axis=1)))
output["lead_jet_eta"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(jet[:, 0:1][BL].eta, axis=1)))
output["lead_jet_phi"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(jet[:, 0:1][BL].phi, axis=1)))
output["sublead_jet_pt"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(jet[:, 1:2][BL].pt, axis=1)))
output["sublead_jet_eta"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(jet[:, 1:2][BL].eta, axis=1)))
output["sublead_jet_phi"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(jet[:, 1:2][BL].phi, axis=1)))
output["lead_btag_pt"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(high_score_btag[:, 0:1][BL].pt, axis=1)))
output["lead_btag_eta"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(high_score_btag[:, 0:1][BL].eta, axis=1)))
output["lead_btag_phi"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(high_score_btag[:, 0:1][BL].phi, axis=1)))
output["sublead_btag_pt"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(high_score_btag[:, 1:2][BL].pt, axis=1)))
output["sublead_btag_eta"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(high_score_btag[:, 1:2][BL].eta, axis=1)))
output["sublead_btag_phi"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(high_score_btag[:, 1:2][BL].phi, axis=1)))
output["fwd_jet_p"] += processor.column_accumulator(
ak.to_numpy(
ak.flatten(ak.fill_none(ak.pad_none(j_fwd[BL].p, 1, clip=True),
0),
axis=1)))
output["fwd_jet_pt"] += processor.column_accumulator(
ak.to_numpy(
ak.flatten(ak.fill_none(
ak.pad_none(j_fwd[BL].pt, 1, clip=True), 0),
axis=1)))
output["fwd_jet_eta"] += processor.column_accumulator(
ak.to_numpy(
ak.flatten(ak.fill_none(
ak.pad_none(j_fwd[BL].eta, 1, clip=True), 0),
axis=1)))
output["fwd_jet_phi"] += processor.column_accumulator(
ak.to_numpy(
ak.flatten(ak.fill_none(
ak.pad_none(j_fwd[BL].phi, 1, clip=True), 0),
axis=1)))
output["mjj_max"] += processor.column_accumulator(
ak.to_numpy(ak.fill_none(ak.max(mjf[BL], axis=1), 0)))
output["delta_eta_jj"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(delta_eta[BL], axis=1)))
output["met"] += processor.column_accumulator(ak.to_numpy(met_pt[BL]))
output["ht"] += processor.column_accumulator(ak.to_numpy(ht[BL]))
output["st"] += processor.column_accumulator(ak.to_numpy(st[BL]))
output["n_jet"] += processor.column_accumulator(
ak.to_numpy(ak.num(jet[BL])))
output["n_btag"] += processor.column_accumulator(
ak.to_numpy(ak.num(btag[BL])))
output["n_fwd"] += processor.column_accumulator(
ak.to_numpy(ak.num(fwd[BL])))
output["n_central"] += processor.column_accumulator(
ak.to_numpy(ak.num(central[BL])))
output["n_tau"] += processor.column_accumulator(
ak.to_numpy(ak.num(tau[BL])))
output["n_track"] += processor.column_accumulator(
ak.to_numpy(ak.num(track[BL])))
output["dilepton_pt"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(dilepton_pt[BL], axis=1)))
output["dilepton_mass"] += processor.column_accumulator(
ak.to_numpy(ak.flatten(dilepton_mass[BL], axis=1)))
output["min_bl_dR"] += processor.column_accumulator(
ak.to_numpy(min_bl_dR[BL]))
output["min_mt_lep_met"] += processor.column_accumulator(
ak.to_numpy(min_mt_lep_met[BL]))
output["label"] += processor.column_accumulator(label)
output["weight"] += processor.column_accumulator(weight.weight()[BL])
output["presel"]["all"] += len(ev[ss_selection])
output["sel"]["all"] += len(ev[BL])
return output
def process(self, events):
output = self.accumulator.identity()
# use a very loose preselection to filter the events
presel = ak.num(events.Jet) > 2
ev = events[presel]
dataset = ev.metadata['dataset']
# load the config - probably not needed anymore
cfg = loadConfig()
output['totalEvents']['all'] += len(events)
output['skimmedEvents']['all'] += len(ev)
## Muons
muon = Collections(ev, "Muon", "tight").get()
vetomuon = Collections(ev, "Muon", "veto").get()
dimuon = choose(muon, 2)
SSmuon = ak.any((dimuon['0'].charge * dimuon['1'].charge) > 0, axis=1)
OSmuon = ak.any((dimuon['0'].charge * dimuon['1'].charge) < 0, axis=1)
leading_muon_idx = ak.singletons(ak.argmax(muon.pt, axis=1))
leading_muon = muon[leading_muon_idx]
## Electrons
electron = Collections(ev, "Electron", "tight").get()
vetoelectron = Collections(ev, "Electron", "veto").get()
dielectron = choose(electron, 2)
SSelectron = ak.any(
(dielectron['0'].charge * dielectron['1'].charge) > 0, axis=1)
OSelectron = ak.any(
(dielectron['0'].charge * dielectron['1'].charge) < 0, axis=1)
leading_electron_idx = ak.singletons(ak.argmax(electron.pt, axis=1))
leading_electron = electron[leading_electron_idx]
## Merge electrons and muons - this should work better now in ak1
lepton = ak.concatenate([muon, electron], axis=1)
dilepton = cross(muon, electron)
SSlepton = ak.any((dilepton['0'].charge * dilepton['1'].charge) > 0,
axis=1)
OSlepton = ak.any((dilepton['0'].charge * dilepton['1'].charge) < 0,
axis=1)
leading_lepton_idx = ak.singletons(ak.argmax(lepton.pt, axis=1))
leading_lepton = lepton[leading_lepton_idx]
trailing_lepton_idx = ak.singletons(ak.argmin(lepton.pt, axis=1))
trailing_lepton = lepton[trailing_lepton_idx]
## Jets
jet = getJets(ev, minPt=25, maxEta=4.7)
jet = jet[(jet.pt > 25) & (jet.jetId > 1)]
jet = jet[~match(jet, muon,
deltaRCut=0.4)] # remove jets that overlap with muons
jet = jet[~match(
jet, electron,
deltaRCut=0.4)] # remove jets that overlap with electrons
## event selectors
filters = getFilters(ev, year=self.year, dataset=dataset)
triggers = getTriggers(ev, year=self.year, dataset=dataset)
dilep = ((ak.num(electron) == 1) & (ak.num(muon) == 1))
lep0pt = ((ak.num(electron[(electron.pt > 25)]) +
ak.num(muon[(muon.pt > 25)])) > 0)
lep1pt = ((ak.num(electron[(electron.pt > 20)]) +
ak.num(muon[(muon.pt > 20)])) > 1)
lepveto = ((ak.num(vetoelectron) + ak.num(vetomuon)) == 2)
# define the weight
weight = Weights(len(ev))
if not dataset == 'MuonEG':
# lumi weight
weight.add("weight", ev.weight * cfg['lumi'][self.year])
## PU weight - not in the babies...
#weight.add("PU", ev.puWeight, weightUp=ev.puWeightUp, weightDown=ev.puWeightDown, shift=False)
# b-tag SFs
#weight.add("btag", self.btagSF.Method1a(btag, light))
# lepton SFs
weight.add("lepton", self.leptonSF.get(electron, muon))
selection = PackedSelection()
selection.add('lepveto', lepveto)
selection.add('dilep', dilep)
selection.add('trigger', (triggers))
selection.add('filter', (filters))
selection.add('p_T(lep0)>25', lep0pt)
selection.add('p_T(lep1)>20', lep1pt)
selection.add('OS', OSlepton)
selection.add('N_jet>2', (ak.num(jet) >= 3))
selection.add('MET>30', (ev.MET.pt > 30))
os_reqs = [
'lepveto', 'dilep', 'trigger', 'filter', 'p_T(lep0)>25',
'p_T(lep1)>20', 'OS'
]
bl_reqs = os_reqs + ['N_jet>2', 'MET>30']
os_reqs_d = {sel: True for sel in os_reqs}
os_selection = selection.require(**os_reqs_d)
bl_reqs_d = {sel: True for sel in bl_reqs}
BL = selection.require(**bl_reqs_d)
cutflow = Cutflow(output, ev, weight=weight)
cutflow_reqs_d = {}
for req in bl_reqs:
cutflow_reqs_d.update({req: True})
cutflow.addRow(req, selection.require(**cutflow_reqs_d))
# first, make a few super inclusive plots
output['PV_npvs'].fill(dataset=dataset,
multiplicity=ev.PV[os_selection].npvs,
weight=weight.weight()[os_selection])
output['PV_npvsGood'].fill(dataset=dataset,
multiplicity=ev.PV[os_selection].npvsGood,
weight=weight.weight()[os_selection])
output['N_jet'].fill(dataset=dataset,
multiplicity=ak.num(jet)[os_selection],
weight=weight.weight()[os_selection])
output['MET'].fill(dataset=dataset,
pt=ev.MET[os_selection].pt,
phi=ev.MET[os_selection].phi,
weight=weight.weight()[os_selection])
output['j1'].fill(dataset=dataset,
pt=ak.flatten(jet.pt[:, 0:1][BL]),
eta=ak.flatten(jet.eta[:, 0:1][BL]),
phi=ak.flatten(jet.phi[:, 0:1][BL]),
weight=weight.weight()[BL])
# Now, take care of systematic unceratinties
if not dataset == 'MuonEG':
alljets = getJets(ev, minPt=0, maxEta=4.7)
alljets = alljets[(alljets.jetId > 1)]
for var in self.variations:
# get the collections that change with the variations
jet_var = getPtEtaPhi(alljets, pt_var=var)
jet_var = jet_var[(jet_var.pt > 25)]
jet_var = jet_var[~match(
jet_var, muon,
deltaRCut=0.4)] # remove jets that overlap with muons
jet_var = jet_var[~match(
jet_var, electron,
deltaRCut=0.4)] # remove jets that overlap with electrons
# get the modified selection -> more difficult
selection.add(
'N_jet>2_' + var, (ak.num(jet_var.pt) > 3)
) # something needs to be improved with getPtEtaPhi function
selection.add('MET>30_' + var, (getattr(ev.MET, var) > 30))
bl_reqs = os_reqs + ['N_jet>2_' + var, 'MET>30_' + var]
bl_reqs_d = {sel: True for sel in bl_reqs}
BL = selection.require(**bl_reqs_d)
# the OS selection remains unchanged
output['N_jet_' + var].fill(
dataset=dataset,
multiplicity=ak.num(jet_var)[os_selection],
weight=weight.weight()[os_selection])
# We don't need to redo all plots with variations. E.g., just add uncertainties to the jet plots.
output['j1_' + var].fill(dataset=dataset,
pt=ak.flatten(jet_var.pt[:, 0:1][BL]),
eta=ak.flatten(jet_var.eta[:,
0:1][BL]),
phi=ak.flatten(jet_var.phi[:,
0:1][BL]),
weight=weight.weight()[BL])
return output
def process(self, events):
output = self.accumulator.identity()
# we can use a very loose preselection to filter the events. nothing is done with this presel, though
presel = ak.num(events.Jet) > 0
ev = events[presel]
dataset = ev.metadata['dataset']
# load the config - probably not needed anymore
cfg = loadConfig()
output['totalEvents']['all'] += len(events)
output['skimmedEvents']['all'] += len(ev)
## Electrons
electron = Collections(ev, "Electron", "tight").get()
electron = electron[(electron.miniPFRelIso_all < 0.12)
& (electron.pt > 20) & (abs(electron.eta) < 2.4)]
gen_matched_electron = electron[((electron.genPartIdx >= 0) & (abs(
electron.matched_gen.pdgId) == 11))]
n_gen = ak.num(gen_matched_electron)
is_flipped = ((gen_matched_electron.matched_gen.pdgId *
(-1) == gen_matched_electron.pdgId) &
(abs(gen_matched_electron.pdgId) == 11))
#is_flipped = (abs(ev.GenPart[gen_matched_electron.genPartIdx].pdgId) == abs(gen_matched_electron.pdgId))&(ev.GenPart[gen_matched_electron.genPartIdx].pdgId/abs(ev.GenPart[gen_matched_electron.genPartIdx].pdgId) != gen_matched_electron.pdgId/abs(gen_matched_electron.pdgId))
flipped_electron = gen_matched_electron[is_flipped]
n_flips = ak.num(flipped_electron)
sielectron = choose(electron, 1)
dielectron = choose(electron, 2)
SSelectron = ak.any(
(dielectron['0'].charge * dielectron['1'].charge) > 0, axis=1)
leading_electron_idx = ak.singletons(ak.argmax(electron.pt, axis=1))
leading_electron = electron[leading_electron_idx]
## MET -> can switch to puppi MET
met_pt = ev.MET.pt
met_phi = ev.MET.phi
# setting up the various weights
weight = Weights(len(ev))
weight2 = Weights(len(ev))
if not dataset == 'MuonEG':
# generator weight
weight.add("weight", ev.genWeight)
weight2.add("weight", ev.genWeight)
weight2.add("charge flip",
self.charge_flip_ratio.flip_ratio(sielectron['0']))
#selections
filters = getFilters(ev, year=self.year, dataset=dataset)
electr = ((ak.num(electron) == 2))
ss = (SSelectron)
gen = (n_gen >= 1)
flip = (n_flips >= 1)
selection = PackedSelection()
selection.add('filter', (filters))
selection.add('electr', electr)
selection.add('ss', ss)
selection.add('flip', flip)
selection.add('gen', gen)
bl_reqs = ['filter', 'electr']
bl_reqs_d = {sel: True for sel in bl_reqs}
baseline = selection.require(**bl_reqs_d)
s_reqs = bl_reqs + ['ss']
s_reqs_d = {sel: True for sel in s_reqs}
ss_sel = selection.require(**s_reqs_d)
f_reqs = bl_reqs + ['gen', 'flip']
f_reqs_d = {sel: True for sel in f_reqs}
flip_sel = selection.require(**f_reqs_d)
#outputs
output['N_ele'].fill(dataset=dataset,
multiplicity=ak.num(electron)[flip_sel],
weight=weight.weight()[flip_sel])
output['electron_flips'].fill(dataset=dataset,
multiplicity=n_flips[flip_sel],
weight=weight.weight()[flip_sel])
output['N_ele2'].fill(dataset=dataset,
multiplicity=ak.num(electron)[baseline],
weight=weight2.weight()[baseline])
output['electron_flips2'].fill(dataset=dataset,
multiplicity=n_flips[baseline],
weight=weight2.weight()[baseline])
output["electron"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(flipped_electron[flip_sel].pt)),
eta=ak.to_numpy(ak.flatten(abs(flipped_electron[flip_sel].eta))),
#phi = ak.to_numpy(ak.flatten(leading_electron[baseline].phi)),
weight=weight.weight()[flip_sel])
output["electron2"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(leading_electron[baseline].pt)),
eta=ak.to_numpy(ak.flatten(abs(leading_electron[baseline].eta))),
#phi = ak.to_numpy(ak.flatten(leading_electron[baseline].phi)),
weight=weight2.weight()[baseline])
return output
def process(self, events):
output = self.accumulator.identity()
# we can use a very loose preselection to filter the events. nothing is done with this presel, though
presel = ak.num(events.Jet) >= 0
ev = events[presel]
dataset = ev.metadata['dataset']
# load the config - probably not needed anymore
cfg = loadConfig()
output['totalEvents']['all'] += len(events)
output['skimmedEvents']['all'] += len(ev)
## Muons
muon = ev.Muon
## Electrons
electron = ev.Electron
## Merge electrons and muons - this should work better now in ak1
dilepton = cross(muon, electron)
SSlepton = ak.any((dilepton['0'].charge * dilepton['1'].charge) > 0,
axis=1)
lepton = ak.concatenate([muon, electron], axis=1)
leading_lepton_idx = ak.singletons(ak.argmax(lepton.pt, axis=1))
leading_lepton = lepton[leading_lepton_idx]
trailing_lepton_idx = ak.singletons(ak.argmin(lepton.pt, axis=1))
trailing_lepton = lepton[trailing_lepton_idx]
## MET -> can switch to puppi MET
met_pt = ev.MET.pt
met_phi = ev.MET.phi
# define the weight
weight = Weights(len(ev))
if not dataset == 'MuonEG':
# generator weight
weight.add("weight", ev.genWeight)
filters = getFilters(ev, year=self.year, dataset=dataset)
dilep = ((ak.num(electron) + ak.num(muon)) == 2)
selection = PackedSelection()
selection.add('dilep', dilep)
selection.add('filter', (filters))
bl_reqs = ['dilep', 'filter']
bl_reqs_d = {sel: True for sel in bl_reqs}
baseline = selection.require(**bl_reqs_d)
output['N_ele'].fill(dataset=dataset,
multiplicity=ak.num(electron)[baseline],
weight=weight.weight()[baseline])
output['N_mu'].fill(dataset=dataset,
multiplicity=ak.num(muon)[baseline],
weight=weight.weight()[baseline])
output['lead_lep'].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(leading_lepton[baseline].pt)),
eta=ak.to_numpy(ak.flatten(leading_lepton[baseline].eta)),
phi=ak.to_numpy(ak.flatten(leading_lepton[baseline].phi)),
weight=weight.weight()[baseline])
return output
def process(self, events):
# Dataset parameters
dataset = events.metadata['dataset']
year = self._samples[dataset]['year']
xsec = self._samples[dataset]['xsec']
sow = self._samples[dataset]['nSumOfWeights' ]
isData = self._samples[dataset]['isData']
datasets = ['SingleMuon', 'SingleElectron', 'EGamma', 'MuonEG', 'DoubleMuon', 'DoubleElectron']
for d in datasets:
if d in dataset: dataset = dataset.split('_')[0]
# Inittialize objects
met = events.GenMET
e = events.GenPart[abs(events.GenPart.pdgId)==11]
mu = events.GenPart[abs(events.GenPart.pdgId)==13]
tau = events.GenPart[abs(events.GenPart.pdgId)==15]
j = events.GenJet
leading_mu = mu[ak.argmax(mu.pt,axis=-1,keepdims=True)]
leading_e = e[ak.argmax(e.pt,axis=-1,keepdims=True)]
nElec = ak.num(e)
nMuon = ak.num(mu)
nTau = ak.num(tau)
twoLeps = (nElec+nMuon) == 2
threeLeps = (nElec+nMuon) == 3
twoElec = (nElec == 2)
twoMuon = (nMuon == 2)
e0 = e[ak.argmax(e.pt,axis=-1,keepdims=True)]
m0 = mu[ak.argmax(mu.pt,axis=-1,keepdims=True)]
elecs = e[ak.argsort(e.pt, ascending=False)]
muons = mu[ak.argsort(mu.pt, ascending=False)]
e1 = elecs
e2 = elecs
m1 = muons
m2 = muons
# Jet selection
jetptname = 'pt_nom' if hasattr(j, 'pt_nom') else 'pt'
njets = ak.num(j)
ht = ak.sum(j.pt,axis=-1)
jets = j[ak.argsort(j.pt, ascending=False)]
j0 = j[ak.argmax(j.pt,axis=-1,keepdims=True)]
j1 = jets
j2 = jets
j3 = jets
nbtags = ak.num(j[abs(j.hadronFlavour)==5])
##################################################################
### 2 same-sign leptons
##################################################################
# emu
singe = e [(nElec==1)&(nMuon==1)&(e .pt>-1)]
singm = mu[(nElec==1)&(nMuon==1)&(mu.pt>-1)]
em = ak.cartesian({"e":singe,"m":singm})
emSSmask = (em.e.pdgId*em.m.pdgId>0)
emSS = em[emSSmask]
nemSS = len(ak.flatten(emSS))
year = 2018
lepSF_emSS = GetLeptonSF(mu.pt, mu.eta, 'm', e.pt, e.eta, 'e', year=year)
# ee and mumu
# pt>-1 to preserve jagged dimensions
ee = e [(nElec==2)&(nMuon==0)&(e.pt>-1)]
mm = mu[(nElec==0)&(nMuon==2)&(mu.pt>-1)]
eepairs = ak.combinations(ee, 2, fields=["e0","e1"])
eeSSmask = (eepairs.e0.pdgId*eepairs.e1.pdgId>0)
eeonZmask = (np.abs((eepairs.e0+eepairs.e1).mass-91.2)<10)
eeoffZmask = (eeonZmask==0)
mmpairs = ak.combinations(mm, 2, fields=["m0","m1"])
mmSSmask = (mmpairs.m0.pdgId*mmpairs.m1.pdgId>0)
mmonZmask = (np.abs((mmpairs.m0+mmpairs.m1).mass-91.2)<10)
mmoffZmask = (mmonZmask==0)
eeSSonZ = eepairs[eeSSmask & eeonZmask]
eeSSoffZ = eepairs[eeSSmask & eeoffZmask]
mmSSonZ = mmpairs[mmSSmask & mmonZmask]
mmSSoffZ = mmpairs[mmSSmask & mmoffZmask]
neeSS = len(ak.flatten(eeSSonZ)) + len(ak.flatten(eeSSoffZ))
nmmSS = len(ak.flatten(mmSSonZ)) + len(ak.flatten(mmSSoffZ))
lepSF_eeSS = GetLeptonSF(eepairs.e0.pt, eepairs.e0.eta, 'e', eepairs.e1.pt, eepairs.e1.eta, 'e', year=year)
lepSF_mumuSS = GetLeptonSF(mmpairs.m0.pt, mmpairs.m0.eta, 'm', mmpairs.m1.pt, mmpairs.m1.eta, 'm', year=year)
print('Same-sign events [ee, emu, mumu] = [%i, %i, %i]'%(neeSS, nemSS, nmmSS))
# Cuts
eeSSmask = (ak.num(eeSSmask[eeSSmask])>0)
mmSSmask = (ak.num(mmSSmask[mmSSmask])>0)
eeonZmask = (ak.num(eeonZmask[eeonZmask])>0)
eeoffZmask = (ak.num(eeoffZmask[eeoffZmask])>0)
mmonZmask = (ak.num(mmonZmask[mmonZmask])>0)
mmoffZmask = (ak.num(mmoffZmask[mmoffZmask])>0)
emSSmask = (ak.num(emSSmask[emSSmask])>0)
##################################################################
### 3 leptons
##################################################################
# eem
muon_eem = mu[(nElec==2)&(nMuon==1)&(mu.pt>-1)]
elec_eem = e[(nElec==2)&(nMuon==1)&( e.pt>-1)]
ee_eem = ak.combinations(elec_eem, 2, fields=["e0", "e1"])
ee_eemZmask = (ee_eem.e0.pdgId*ee_eem.e1.pdgId<1)&(np.abs((ee_eem.e0+ee_eem.e1).mass-91.2)<10)
ee_eemOffZmask = (ee_eem.e0.pdgId*ee_eem.e1.pdgId<1)&(np.abs((ee_eem.e0+ee_eem.e1).mass-91.2)>10)
ee_eemZmask = (ak.num(ee_eemZmask[ee_eemZmask])>0)
ee_eemOffZmask = (ak.num(ee_eemOffZmask[ee_eemOffZmask])>0)
eepair_eem = (ee_eem.e0+ee_eem.e1)
trilep_eem = eepair_eem+muon_eem #ak.cartesian({"e0":ee_eem.e0,"e1":ee_eem.e1, "m":muon_eem})
lepSF_eem = GetLeptonSF(ee_eem.e0.pt, ee_eem.e0.eta, 'e', ee_eem.e1.pt, ee_eem.e1.eta, 'e', mu.pt, mu.eta, 'm', year)
# mme
muon_mme = mu[(nElec==1)&(nMuon==2)&(mu.pt>-1)]
elec_mme = e[(nElec==1)&(nMuon==2)&( e.pt>-1)]
mm_mme = ak.combinations(muon_mme, 2, fields=["m0", "m1"])
mm_mmeZmask = (mm_mme.m0.pdgId*mm_mme.m1.pdgId<1)&(np.abs((mm_mme.m0+mm_mme.m1).mass-91.2)<10)
mm_mmeOffZmask = (mm_mme.m0.pdgId*mm_mme.m1.pdgId<1)&(np.abs((mm_mme.m0+mm_mme.m1).mass-91.2)>10)
mm_mmeZmask = (ak.num(mm_mmeZmask[mm_mmeZmask])>0)
mm_mmeOffZmask = (ak.num(mm_mmeOffZmask[mm_mmeOffZmask])>0)
mmpair_mme = (mm_mme.m0+mm_mme.m1)
trilep_mme = mmpair_mme+elec_mme
mZ_mme = mmpair_mme.mass
mZ_eem = eepair_eem.mass
m3l_eem = trilep_eem.mass
m3l_mme = trilep_mme.mass
lepSF_mme = GetLeptonSF(mm_mme.m0.pt, mm_mme.m0.eta, 'm', mm_mme.m1.pt, mm_mme.m1.eta, 'm', e.pt, e.eta, 'e', year)
# eee and mmm
eee = e[(nElec==3)&(nMuon==0)&( e.pt>-1)]
mmm = mu[(nElec==0)&(nMuon==3)&(mu.pt>-1)]
eee_leps = ak.combinations(eee, 3, fields=["e0", "e1", "e2"])
mmm_leps = ak.combinations(mmm, 3, fields=["m0", "m1", "m2"])
ee_pairs = ak.combinations(eee, 2, fields=["e0", "e1"])
mm_pairs = ak.combinations(mmm, 2, fields=["m0", "m1"])
ee_pairs_index = ak.argcombinations(eee, 2, fields=["e0", "e1"])
mm_pairs_index = ak.argcombinations(mmm, 2, fields=["m0", "m1"])
lepSF_eee = GetLeptonSF(eee_leps.e0.pt, eee_leps.e0.eta, 'e', eee_leps.e1.pt, eee_leps.e1.eta, 'e', eee_leps.e2.pt, eee_leps.e2.eta, 'e', year)
lepSF_mmm = GetLeptonSF(mmm_leps.m0.pt, mmm_leps.m0.eta, 'm', mmm_leps.m1.pt, mmm_leps.m1.eta, 'm', mmm_leps.m2.pt, mmm_leps.m2.eta, 'm', year)
mmSFOS_pairs = mm_pairs[(np.abs(mm_pairs.m0.pdgId) == np.abs(mm_pairs.m1.pdgId)) & (mm_pairs.m0.pdgId != mm_pairs.m1.pdgId)]
offZmask_mm = ak.all(np.abs((mmSFOS_pairs.m0 + mmSFOS_pairs.m1).mass - 91.2)>10., axis=1, keepdims=True) & (ak.num(mmSFOS_pairs)>0)
onZmask_mm = ak.any(np.abs((mmSFOS_pairs.m0 + mmSFOS_pairs.m1).mass - 91.2)<10., axis=1, keepdims=True)
eeSFOS_pairs = ee_pairs[(np.abs(ee_pairs.e0.pdgId) == np.abs(ee_pairs.e1.pdgId)) & (ee_pairs.e0.pdgId != ee_pairs.e1.pdgId)]
offZmask_ee = ak.all(np.abs((eeSFOS_pairs.e0 + eeSFOS_pairs.e1).mass - 91.2)>10, axis=1, keepdims=True) & (ak.num(eeSFOS_pairs)>0)
onZmask_ee = ak.any(np.abs((eeSFOS_pairs.e0 + eeSFOS_pairs.e1).mass - 91.2)<10, axis=1, keepdims=True)
# Create masks **for event selection**
eeeOnZmask = (ak.num(onZmask_ee[onZmask_ee])>0)
eeeOffZmask = (ak.num(offZmask_ee[offZmask_ee])>0)
mmmOnZmask = (ak.num(onZmask_mm[onZmask_mm])>0)
mmmOffZmask = (ak.num(offZmask_mm[offZmask_mm])>0)
# Now we need to create masks for the leptons in order to select leptons from the Z boson candidate (in onZ categories)
ZeeMask = ak.argmin(np.abs((eeSFOS_pairs.e0 + eeSFOS_pairs.e1).mass - 91.2),axis=1,keepdims=True)
ZmmMask = ak.argmin(np.abs((mmSFOS_pairs.m0 + mmSFOS_pairs.m1).mass - 91.2),axis=1,keepdims=True)
Zee = eeSFOS_pairs[ZeeMask]
Zmm = mmSFOS_pairs[ZmmMask]
eZ0= Zee.e0[ak.num(eeSFOS_pairs)>0]
eZ1= Zee.e1[ak.num(eeSFOS_pairs)>0]
eZ = eZ0+eZ1
mZ0= Zmm.m0[ak.num(mmSFOS_pairs)>0]
mZ1= Zmm.m1[ak.num(mmSFOS_pairs)>0]
mZ = mZ0+mZ1
mZ_eee = eZ.mass
mZ_mmm = mZ.mass
# And for the W boson
ZmmIndices = mm_pairs_index[ZmmMask]
ZeeIndices = ee_pairs_index[ZeeMask]
eW = eee[~ZeeIndices.e0 | ~ZeeIndices.e1]
mW = mmm[~ZmmIndices.m0 | ~ZmmIndices.m1]
triElec = eee_leps.e0+eee_leps.e1+eee_leps.e2
triMuon = mmm_leps.m0+mmm_leps.m1+mmm_leps.m2
m3l_eee = triElec.mass
m3l_mmm = triMuon.mass
# Triggers
trig_eeSS = passTrigger(events,'ee',isData,dataset)
trig_mmSS = passTrigger(events,'mm',isData,dataset)
trig_emSS = passTrigger(events,'em',isData,dataset)
trig_eee = passTrigger(events,'eee',isData,dataset)
trig_mmm = passTrigger(events,'mmm',isData,dataset)
trig_eem = passTrigger(events,'eem',isData,dataset)
trig_mme = passTrigger(events,'mme',isData,dataset)
# MET filters
# Weights
genw = np.ones_like(events['MET_pt']) if isData else events['genWeight']
### We need weights for: normalization, lepSF, triggerSF, pileup, btagSF...
weights = {}
for r in ['all', 'ee', 'mm', 'em', 'eee', 'mmm', 'eem', 'mme']:
weights[r] = coffea.analysis_tools.Weights(len(events))
weights[r].add('norm',genw if isData else (xsec/sow)*genw)
weights['ee'].add('lepSF_eeSS', lepSF_eeSS)
weights['em'].add('lepSF_emSS', lepSF_emSS)
weights['mm'].add('lepSF_mmSS', lepSF_mumuSS)
weights['eee'].add('lepSF_eee', lepSF_eee)
weights['mmm'].add('lepSF_mmm', lepSF_mmm)
weights['mme'].add('lepSF_mme', lepSF_mme)
weights['eem'].add('lepSF_eem', lepSF_eem)
# Extract the EFT quadratic coefficients and optionally use them to calculate the coefficients on the w**2 quartic function
# eft_coeffs is never Jagged so convert immediately to numpy for ease of use.
eft_coeffs = ak.to_numpy(events['EFTfitCoefficients']) if hasattr(events, "EFTfitCoefficients") else None
eft_w2_coeffs = efth.calc_w2_coeffs(eft_coeffs,self._dtype) if (self._do_errors and eft_coeffs is not None) else None
# Selections and cuts
selections = PackedSelection()
channels2LSS = ['eeSSonZ', 'eeSSoffZ', 'mmSSonZ', 'mmSSoffZ', 'emSS']
selections.add('eeSSonZ', (eeonZmask)&(eeSSmask)&(trig_eeSS))
selections.add('eeSSoffZ', (eeoffZmask)&(eeSSmask)&(trig_eeSS))
selections.add('mmSSonZ', (mmonZmask)&(mmSSmask)&(trig_mmSS))
selections.add('mmSSoffZ', (mmoffZmask)&(mmSSmask)&(trig_mmSS))
selections.add('emSS', (emSSmask)&(trig_emSS))
channels3L = ['eemSSonZ', 'eemSSoffZ', 'mmeSSonZ', 'mmeSSoffZ']
selections.add('eemSSonZ', (ee_eemZmask)&(trig_eem))
selections.add('eemSSoffZ', (ee_eemOffZmask)&(trig_eem))
selections.add('mmeSSonZ', (mm_mmeZmask)&(trig_mme))
selections.add('mmeSSoffZ', (mm_mmeOffZmask)&(trig_mme))
channels3L += ['eeeSSonZ', 'eeeSSoffZ', 'mmmSSonZ', 'mmmSSoffZ']
selections.add('eeeSSonZ', (eeeOnZmask)&(trig_eee))
selections.add('eeeSSoffZ', (eeeOffZmask)&(trig_eee))
selections.add('mmmSSonZ', (mmmOnZmask)&(trig_mmm))
selections.add('mmmSSoffZ', (mmmOffZmask)&(trig_mmm))
levels = ['base', '2jets', '4jets', '4j1b', '4j2b']
selections.add('base', (nElec+nMuon>=2))
selections.add('2jets',(njets>=2))
selections.add('4jets',(njets>=4))
selections.add('4j1b',(njets>=4)&(nbtags>=1))
selections.add('4j2b',(njets>=4)&(nbtags>=2))
# Variables
invMass_eeSSonZ = ( eeSSonZ.e0+ eeSSonZ.e1).mass
invMass_eeSSoffZ = (eeSSoffZ.e0+eeSSoffZ.e1).mass
invMass_mmSSonZ = ( mmSSonZ.m0+ mmSSonZ.m1).mass
invMass_mmSSoffZ = (mmSSoffZ.m0+mmSSoffZ.m1).mass
invMass_emSS = (emSS.e+emSS.m).mass
varnames = {}
varnames['met'] = met.pt
varnames['ht'] = ht
varnames['njets'] = njets
varnames['nbtags'] = nbtags
varnames['invmass'] = {
'eeSSonZ' : invMass_eeSSonZ,
'eeSSoffZ' : invMass_eeSSoffZ,
'mmSSonZ' : invMass_mmSSonZ,
'mmSSoffZ' : invMass_mmSSoffZ,
'emSS' : invMass_emSS,
'eemSSonZ' : mZ_eem,
'eemSSoffZ' : mZ_eem,
'mmeSSonZ' : mZ_mme,
'mmeSSoffZ' : mZ_mme,
'eeeSSonZ' : mZ_eee,
'eeeSSoffZ' : mZ_eee,
'mmmSSonZ' : mZ_mmm,
'mmmSSoffZ' : mZ_mmm,
}
varnames['m3l'] = {
'eemSSonZ' : m3l_eem,
'eemSSoffZ' : m3l_eem,
'mmeSSonZ' : m3l_mme,
'mmeSSoffZ' : m3l_mme,
'eeeSSonZ' : m3l_eee,
'eeeSSoffZ' : m3l_eee,
'mmmSSonZ' : m3l_mmm,
'mmmSSoffZ' : m3l_mmm,
}
varnames['e0pt' ] = e0.pt
varnames['e0eta'] = e0.eta
varnames['m0pt' ] = m0.pt
varnames['m0eta'] = m0.eta
varnames['e1pt' ] = e1
varnames['e1eta'] = e1
varnames['e2pt' ] = e2
varnames['e2eta'] = e2
varnames['m1pt' ] = m1
varnames['m1eta'] = m1
varnames['m2pt' ] = m2
varnames['m2eta'] = m2
varnames['j0pt' ] = j0.pt
varnames['j0eta'] = j0.eta
varnames['j1pt'] = j1
varnames['j1eta'] = j1
varnames['j2pt'] = j2
varnames['j2eta'] = j2
varnames['j3pt'] = j3
varnames['j3eta'] = j3
varnames['counts'] = np.ones_like(events.GenMET.pt)
# fill Histos
hout = self.accumulator.identity()
normweights = weights['all'].weight().flatten() # Why does it not complain about .flatten() here?
hout['SumOfEFTweights'].fill(sample=dataset, SumOfEFTweights=varnames['counts'], weight=normweights, eft_coeff=eft_coeffs, eft_err_coeff=eft_w2_coeffs)
for var, v in varnames.items():
for ch in channels2LSS+channels3L:
for lev in levels:
weight = weights[ ch[:3] if (ch.startswith('eee') or ch.startswith('mmm') or ch.startswith('eem') or ch.startswith('mme')) else ch[:2]].weight()
cuts = [ch] + [lev]
cut = selections.all(*cuts)
weights_flat = weight[cut].flatten() # Why does it not complain about .flatten() here?
weights_ones = np.ones_like(weights_flat, dtype=np.int)
eft_coeffs_cut = eft_coeffs[cut] if eft_coeffs is not None else None
eft_w2_coeffs_cut = eft_w2_coeffs[cut] if eft_w2_coeffs is not None else None
if var == 'invmass':
if ch in ['eeeSSoffZ', 'mmmSSoffZ']: continue
elif ch in ['eeeSSonZ' , 'mmmSSonZ' ]: continue #values = v[ch]
else : values = ak.flatten(v[ch][cut])
hout['invmass'].fill(eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut, sample=dataset, channel=ch, cut=lev, invmass=values, weight=weights_flat)
elif var == 'm3l':
if ch in ['eeSSonZ','eeSSoffZ', 'mmSSonZ', 'mmSSoffZ','emSS', 'eeeSSoffZ', 'mmmSSoffZ', 'eeeSSonZ' , 'mmmSSonZ']: continue
values = ak.flatten(v[ch][cut])
hout['m3l'].fill(sample=dataset, channel=ch, cut=lev, m3l=values, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
else:
values = v[cut]
if var == 'ht' : hout[var].fill(ht=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'met' : hout[var].fill(met=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'njets' : hout[var].fill(njets=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'nbtags': hout[var].fill(nbtags=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'counts': hout[var].fill(counts=values, sample=dataset, channel=ch, cut=lev, weight=weights_ones)
elif var == 'j0eta' :
if lev == 'base': continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(j0eta=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'e0pt' :
if ch in ['mmSSonZ', 'mmSSoffZ', 'mmmSSoffZ', 'mmmSSonZ']: continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(e0pt=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'm0pt' :
if ch in ['eeSSonZ', 'eeSSoffZ', 'eeeSSoffZ', 'eeeSSonZ']: continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(m0pt=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'e0eta' :
if ch in ['mmSSonZ', 'mmSSoffZ', 'mmmSSoffZ', 'mmmSSonZ']: continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(e0eta=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'm0eta':
if ch in ['eeSSonZ', 'eeSSoffZ', 'eeeSSoffZ', 'eeeSSonZ']: continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(m0eta=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'j0pt' :
if lev == 'base': continue
values = ak.flatten(values)
#values=np.asarray(values)
hout[var].fill(j0pt=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'j1pt':
if lev == "base": continue
values = values.pt[:,1]
#values = ak.flatten(values)
hout[var].fill(j1pt=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var =='j1eta':
if lev == 'base': continue
values = values.eta[:,1]
hout[var].fill(j1eta=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'j2pt':
if lev in ['base', "2jets"]: continue
values = values.pt[:,2]
hout[var].fill(j2pt=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'j2eta':
if lev in ['base', "2jets"]: continue
values = values.eta[:,2]
hout[var].fill(j2eta=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'j3pt':
if lev in ['base', "2jets"]: continue
values = values.pt[:,3]
hout[var].fill(j3pt=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'j3eta':
if lev in ['base', "2jets"]: continue
values = values.eta[:,3]
hout[var].fill(j3eta=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'e1pt':
if ch in ['mmSSonZ', 'mmSSoffZ', 'mmmSSoffZ', 'mmmSSonZ', 'mmeSSonZ', 'mmeSSoffZ', 'emSS']: continue
values = values.pt[:,1]
hout[var].fill(e1pt=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'e1eta':
if ch in ['mmSSonZ', 'mmSSoffZ', 'mmmSSoffZ', 'mmmSSonZ', 'mmeSSonZ', 'mmeSSoffZ', 'emSS']: continue
values = values.eta[:,1]
hout[var].fill(e1eta=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'e2pt':
if ch in ['eeeSSonZ', 'eeeSSoffZ']:
values = values.pt[:,2]
hout[var].fill(e2pt=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'e2eta':
if ch in ['eeeSSonZ', 'eeeSSoffZ']:
values = values.eta[:,2]
hout[var].fill(e2eta=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'm1pt':
if ch in ['eeSSonZ', 'eeSSoffZ', 'eeeSSoffZ', 'eeeSSonZ', 'eemSSonZ', 'eemSSoffZ', 'emSS']: continue
values = values.pt[:,1]
hout[var].fill(m1pt=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'm1eta':
if ch in ['eeSSonZ', 'eeSSoffZ', 'eeeSSoffZ', 'eeeSSonZ', 'eemSSonZ', 'eemSSoffZ', 'emSS']: continue
values = values.eta[:,1]
hout[var].fill(m1eta=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'm2pt':
if ch in ['mmmSSonZ', 'mmmSSoffZ']:
values = values.pt[:,2]
hout[var].fill(m2pt=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
elif var == 'm2eta':
if ch in ['mmmSSonZ', 'mmmSSoffZ']:
values = values.eta[:,2]
hout[var].fill(m2eta=values, sample=dataset, channel=ch, cut=lev, weight=weights_flat, eft_coeff=eft_coeffs_cut, eft_err_coeff=eft_w2_coeffs_cut)
return hout
def process(self, events):
dataset = events.metadata['dataset']
isRealData = not hasattr(events, "genWeight")
selection = PackedSelection()
weights = Weights(len(events))
output = self.accumulator.identity()
if not isRealData:
output['sumw'][dataset] += ak.sum(events.genWeight)
if isRealData:
trigger = np.zeros(len(events), dtype='bool')
for t in self._triggers[self._year]:
trigger = trigger | events.HLT[t]
else:
trigger = np.ones(len(events), dtype='bool')
selection.add('trigger', trigger)
if isRealData:
trigger = np.zeros(len(events), dtype='bool')
for t in self._muontriggers[self._year]:
trigger = trigger | events.HLT[t]
else:
trigger = np.ones(len(events), dtype='bool')
selection.add('muontrigger', trigger)
fatjets = events.FatJet
fatjets['msdcorr'] = corrected_msoftdrop(fatjets)
fatjets['qcdrho'] = 2 * np.log(fatjets.msdcorr / fatjets.pt)
fatjets['n2ddt'] = fatjets.n2b1 - n2ddt_shift(fatjets, year=self._year)
fatjets['msdcorr_full'] = fatjets['msdcorr'] * self._msdSF[self._year]
candidatejet = fatjets[
# https://github.com/DAZSLE/BaconAnalyzer/blob/master/Analyzer/src/VJetLoader.cc#L269
(fatjets.pt > 200)
& (abs(fatjets.eta) < 2.5)
& fatjets.isTight # this is loose in sampleContainer
]
if self._jet_arbitration == 'pt':
candidatejet = ak.firsts(candidatejet)
elif self._jet_arbitration == 'mass':
candidatejet = candidatejet[ak.argmax(candidatejet.msdcorr)]
elif self._jet_arbitration == 'n2':
candidatejet = candidatejet[ak.argmin(candidatejet.n2ddt)]
elif self._jet_arbitration == 'ddb':
candidatejet = candidatejet[ak.argmax(candidatejet.btagDDBvL)]
else:
raise RuntimeError("Unknown candidate jet arbitration")
selection.add('minjetkin', (candidatejet.pt >= 450)
& (candidatejet.msdcorr >= 40.)
& (abs(candidatejet.eta) < 2.5))
selection.add('jetacceptance', (candidatejet.msdcorr >= 47.)
& (candidatejet.pt < 1200)
& (candidatejet.msdcorr < 201.))
selection.add('jetid', candidatejet.isTight)
selection.add('n2ddt', (candidatejet.n2ddt < 0.))
selection.add('ddbpass', (candidatejet.btagDDBvL >= 0.89))
jets = events.Jet[(events.Jet.pt > 30.)
& (abs(events.Jet.eta) < 2.5)
& events.Jet.isTight]
# only consider first 4 jets to be consistent with old framework
jets = jets[:, :4]
dphi = abs(jets.delta_phi(candidatejet))
selection.add(
'antiak4btagMediumOppHem',
ak.max(
jets[dphi > np.pi / 2].btagDeepB, axis=1, mask_identity=False)
< BTagEfficiency.btagWPs[self._year]['medium'])
ak4_away = jets[dphi > 0.8]
selection.add(
'ak4btagMedium08',
ak.max(ak4_away.btagDeepB, axis=1, mask_identity=False) >
BTagEfficiency.btagWPs[self._year]['medium'])
selection.add('met', events.MET.pt < 140.)
goodmuon = ((events.Muon.pt > 10)
& (abs(events.Muon.eta) < 2.4)
& (events.Muon.pfRelIso04_all < 0.25)
& events.Muon.looseId)
nmuons = ak.sum(goodmuon, axis=1)
leadingmuon = ak.firsts(events.Muon[goodmuon])
nelectrons = ak.sum(
(events.Electron.pt > 10)
& (abs(events.Electron.eta) < 2.5)
& (events.Electron.cutBased >= events.Electron.LOOSE),
axis=1,
)
ntaus = ak.sum(
(events.Tau.pt > 20)
& events.Tau.idDecayMode, # bacon iso looser than Nano selection
axis=1,
)
selection.add('noleptons',
(nmuons == 0) & (nelectrons == 0) & (ntaus == 0))
selection.add('onemuon',
(nmuons == 1) & (nelectrons == 0) & (ntaus == 0))
selection.add('muonkin',
(leadingmuon.pt > 55.) & (abs(leadingmuon.eta) < 2.1))
selection.add('muonDphiAK8',
abs(leadingmuon.delta_phi(candidatejet)) > 2 * np.pi / 3)
if isRealData:
genflavor = 0
else:
weights.add('genweight', events.genWeight)
add_pileup_weight(weights, events.Pileup.nPU, self._year, dataset)
bosons = getBosons(events.GenPart)
matchedBoson = candidatejet.nearest(bosons,
axis=None,
threshold=0.8)
genflavor = bosonFlavor(matchedBoson)
genBosonPt = ak.fill_none(ak.firsts(bosons.pt), 0)
add_VJets_NLOkFactor(weights, genBosonPt, self._year, dataset)
add_jetTriggerWeight(weights, candidatejet.msdcorr,
candidatejet.pt, self._year)
output['btagWeight'].fill(dataset=dataset,
val=self._btagSF.addBtagWeight(
weights, ak4_away))
logger.debug("Weight statistics: %r" % weights.weightStatistics)
msd_matched = candidatejet.msdcorr * self._msdSF[self._year] * (
genflavor > 0) + candidatejet.msdcorr * (genflavor == 0)
regions = {
'signal': [
'trigger', 'minjetkin', 'jetacceptance', 'jetid', 'n2ddt',
'antiak4btagMediumOppHem', 'met', 'noleptons'
],
'muoncontrol': [
'muontrigger', 'minjetkin', 'jetacceptance', 'jetid', 'n2ddt',
'ak4btagMedium08', 'onemuon', 'muonkin', 'muonDphiAK8'
],
'noselection': [],
}
for region, cuts in regions.items():
allcuts = set()
output['cutflow'].fill(dataset=dataset,
region=region,
genflavor=genflavor,
cut=0,
weight=weights.weight())
for i, cut in enumerate(cuts + ['ddbpass']):
allcuts.add(cut)
cut = selection.all(*allcuts)
output['cutflow'].fill(dataset=dataset,
region=region,
genflavor=genflavor[cut],
cut=i + 1,
weight=weights.weight()[cut])
systematics = [
None,
'jet_triggerUp',
'jet_triggerDown',
'btagWeightUp',
'btagWeightDown',
'btagEffStatUp',
'btagEffStatDown',
]
def normalize(val, cut):
return ak.to_numpy(ak.fill_none(val[cut], np.nan))
def fill(region, systematic, wmod=None):
selections = regions[region]
cut = selection.all(*selections)
sname = 'nominal' if systematic is None else systematic
if wmod is None:
weight = weights.weight(modifier=systematic)[cut]
else:
weight = weights.weight()[cut] * wmod[cut]
output['templates'].fill(
dataset=dataset,
region=region,
systematic=sname,
genflavor=genflavor[cut],
pt=normalize(candidatejet.pt, cut),
msd=normalize(msd_matched, cut),
ddb=normalize(candidatejet.btagDDBvL, cut),
weight=weight,
)
if wmod is not None:
output['genresponse_noweight'].fill(
dataset=dataset,
region=region,
systematic=sname,
pt=normalize(candidatejet.pt, cut),
genpt=normalize(genBosonPt, cut),
weight=events.genWeight[cut] * wmod[cut],
)
output['genresponse'].fill(
dataset=dataset,
region=region,
systematic=sname,
pt=normalize(candidatejet.pt, cut),
genpt=normalize(genBosonPt, cut),
weight=weight,
)
for region in regions:
cut = selection.all(*(set(regions[region]) - {'n2ddt'}))
output['nminus1_n2ddt'].fill(
dataset=dataset,
region=region,
n2ddt=normalize(candidatejet.n2ddt, cut),
weight=weights.weight()[cut],
)
for systematic in systematics:
fill(region, systematic)
if 'GluGluHToBB' in dataset:
for i in range(9):
fill(region, 'LHEScale_%d' % i, events.LHEScaleWeight[:,
i])
for c in events.LHEWeight.columns[1:]:
fill(region, 'LHEWeight_%s' % c, events.LHEWeight[c])
output["weightStats"] = weights.weightStatistics
return output
def process(self, events):
np.random.seed(
10
) # sets seed so values from random distributions are reproducible (JER corrections)
output = self.accumulator.identity()
self.sample_name = events.metadata['dataset']
## make event weights
# data or MC distinction made internally
mu_evt_weights = MCWeights.get_event_weights(
events, year=args.year, corrections=self.corrections)
el_evt_weights = MCWeights.get_event_weights(
events, year=args.year, corrections=self.corrections)
## initialize selections
selection = PackedSelection()
# get all passing leptons
lep_and_filter_pass = objsel.select_leptons(
events, year=args.year) #, cutflow=output['cutflow'])
selection.add('lep_and_filter_pass', lep_and_filter_pass
) # add passing leptons requirement to all systematics
## build corrected jets and MET
events['Jet'], events['MET'] = IDJet.process_jets(
events, args.year, self.corrections['JetCor'])
runs = events.run
lumis = events.luminosityBlock
Golden_Json_LumiMask = lumi_tools.LumiMask(lumiMask_path)
LumiMask = Golden_Json_LumiMask.__call__(
runs, lumis) ## returns array of valid events
selection.add('lumimask', LumiMask)
## object selection and add different selections
if isSM_Data_:
## muons
selection.add(
'tight_MU',
ak.sum(events['Muon']['TIGHTMU'],
axis=1) == 1) # one muon passing TIGHT criteria
selection.add(
'loose_MU',
ak.sum(events['Muon']['LOOSEMU'],
axis=1) == 1) # one muon passing LOOSE criteria
if isSE_Data_:
## electrons
selection.add(
'tight_EL',
ak.sum(events['Electron']['TIGHTEL'],
axis=1) == 1) # one electron passing TIGHT criteria
selection.add(
'loose_EL',
ak.sum(events['Electron']['LOOSEEL'],
axis=1) == 1) # one electron passing LOOSE criteria
output['cutflow']['lep_and_filter_pass'] += ak.sum(lep_and_filter_pass)
# jet selection
passing_jets = objsel.jets_selection(events,
year=args.year,
cutflow=output['cutflow'])
output['cutflow'][
'nEvts passing jet and lepton obj selection'] += ak.sum(
passing_jets & lep_and_filter_pass)
selection.add('passing_jets', passing_jets)
selection.add('jets_3', ak.num(events['SelectedJets']) == 3)
selection.add('jets_4p',
ak.num(events['SelectedJets']) >
3) # only for getting btag weights
selection.add('DeepCSV_pass',
ak.sum(events['SelectedJets'][btag_wps[0]], axis=1) >= 2)
# sort jets by btag value
events['SelectedJets'] = events['SelectedJets'][ak.argsort(
events['SelectedJets']['btagDeepB'], ascending=False
)] if btaggers[0] == 'DeepCSV' else events['SelectedJets'][ak.argsort(
events['SelectedJets']['btagDeepFlavB'], ascending=False)]
# btag fail sideband
deepcsv_sorted = events['SelectedJets'][ak.argsort(
events['SelectedJets']['btagDeepB'], ascending=False)]['btagDeepB']
valid_counts_inds = ak.where(ak.num(events['SelectedJets']) > 1)[0]
deepcsv_fail = np.zeros(len(events)).astype(bool)
deepcsv_fail[valid_counts_inds] = (
deepcsv_sorted[valid_counts_inds][:, 0] <
btag_values[args.year]['btagDeepB']['DeepCSV' + wps_to_use[0]]) & (
deepcsv_sorted[valid_counts_inds][:, 1] <
btag_values[args.year]['btagDeepB']['DeepCSV' + wps_to_use[0]])
selection.add(
'DeepCSV_fail', deepcsv_fail
) # highest and second highest DeepCSV values don't pass tight and loose WPs
## fill hists for each region
for lepton in self.regions.keys():
evt_weights = mu_evt_weights if lepton == 'Muon' else el_evt_weights
for leptype in self.regions[lepton].keys():
for btagregion in self.regions[lepton][leptype].keys():
for jmult in self.regions[lepton][leptype][
btagregion].keys():
cut = selection.all(
*self.regions[lepton][leptype][btagregion][jmult])
#set_trace()
output['cutflow']['nEvts %s' % ', '.join([
lepton, leptype, btagregion, jmult
])] += cut.sum()
if to_debug: print(lepton, leptype, btagregion, jmult)
if cut.sum() > 0:
ltype = 'MU' if lepton == 'Muon' else 'EL'
if 'loose_or_tight_%s' % ltype in self.regions[
lepton][leptype][btagregion][jmult]:
leptons = events[lepton][cut][(
(events[lepton][cut]['TIGHT%s' % ltype]
== True) |
(events[lepton][cut]['LOOSE%s' % ltype]
== True))]
elif 'tight_%s' % ltype in self.regions[lepton][
leptype][btagregion][jmult]:
leptons = events[lepton][cut][(
events[lepton][cut]['TIGHT%s' %
ltype] == True)]
elif 'loose_%s' % ltype in self.regions[lepton][
leptype][btagregion][jmult]:
leptons = events[lepton][cut][(
events[lepton][cut]['LOOSE%s' %
ltype] == True)]
else:
raise ValueError(
"Not sure what lepton type to choose for event"
)
# get jets and MET
jets, met = events['SelectedJets'][cut], events[
'SelectedMET'][cut]
# find best permutations
best_perms = ttpermutator.find_best_permutations(
jets=jets,
leptons=leptons,
MET=met,
btagWP=btag_wps[0],
btag_req=False
if btagregion == 'btagFail' else True)
valid_perms = ak.num(best_perms['TTbar'].pt) > 0
output['cutflow'][
'nEvts %s: valid perms' %
', '.join([lepton, leptype, btagregion, jmult
])] += ak.sum(valid_perms)
bp_status = np.zeros(
cut.size, dtype=int
) # 0 == '' (no gen matching), 1 == 'right', 2 == 'matchable', 3 == 'unmatchable', 4 == 'sl_tau', 5 == 'noslep'
## create MT regions
MT = make_vars.MT(leptons, met)
MTHigh = ak.flatten(MT[valid_perms] >= MTcut)
output['cutflow'][
'nEvts %s: pass MT cut' %
', '.join([lepton, leptype, btagregion, jmult
])] += ak.sum(MTHigh)
# fill hists for before/after HEM issue
wts = evt_weights.weight(
)[cut][valid_perms][MTHigh]
valid_runs = runs[cut][valid_perms][MTHigh]
runs_before = ak.where(valid_runs < hem_run_after)
runs_after = ak.where(valid_runs >= hem_run_after)
if len(runs_before[0]) > 0:
output = self.fill_hists(
acc=output,
hem_reg='Before',
jetmult=jmult,
leptype=lepton,
lepcat=leptype,
btagregion=btagregion,
permarray=bp_status[cut][valid_perms]
[MTHigh][runs_before],
perm=best_perms[valid_perms][MTHigh]
[runs_before],
jets=jets[valid_perms][MTHigh]
[runs_before],
leptons=leptons[valid_perms][MTHigh]
[runs_before],
MTvals=MT[valid_perms][MTHigh]
[runs_before],
evt_wts=wts[runs_before])
if len(runs_after[0]) > 0:
output = self.fill_hists(
acc=output,
hem_reg='After',
jetmult=jmult,
leptype=lepton,
lepcat=leptype,
btagregion=btagregion,
permarray=bp_status[cut][valid_perms]
[MTHigh][runs_after],
perm=best_perms[valid_perms][MTHigh]
[runs_after],
jets=jets[valid_perms][MTHigh][runs_after],
leptons=leptons[valid_perms][MTHigh]
[runs_after],
MTvals=MT[valid_perms][MTHigh][runs_after],
evt_wts=wts[runs_after])
return output
def process(self, events):
output = self.accumulator.identity()
# use a very loose preselection to filter the events
presel = ak.num(events.Jet)>2
ev = events[presel]
dataset = ev.metadata['dataset']
# load the config - probably not needed anymore
cfg = loadConfig()
output['totalEvents']['all'] += len(events)
output['skimmedEvents']['all'] += len(ev)
## Generated leptons
gen_lep = ev.GenL
leading_gen_lep = gen_lep[ak.singletons(ak.argmax(gen_lep.pt, axis=1))]
trailing_gen_lep = gen_lep[ak.singletons(ak.argmin(gen_lep.pt, axis=1))]
## Muons
muon = Collections(ev, "Muon", "tightSSTTH").get()
vetomuon = Collections(ev, "Muon", "vetoTTH").get()
dimuon = choose(muon, 2)
SSmuon = ak.any((dimuon['0'].charge * dimuon['1'].charge)>0, axis=1)
leading_muon_idx = ak.singletons(ak.argmax(muon.pt, axis=1))
leading_muon = muon[leading_muon_idx]
## Electrons
electron = Collections(ev, "Electron", "tightSSTTH").get()
vetoelectron = Collections(ev, "Electron", "vetoTTH").get()
dielectron = choose(electron, 2)
SSelectron = ak.any((dielectron['0'].charge * dielectron['1'].charge)>0, axis=1)
leading_electron_idx = ak.singletons(ak.argmax(electron.pt, axis=1))
leading_electron = electron[leading_electron_idx]
## Merge electrons and muons - this should work better now in ak1
dilepton = cross(muon, electron)
SSlepton = ak.any((dilepton['0'].charge * dilepton['1'].charge)>0, axis=1)
lepton = ak.concatenate([muon, electron], axis=1)
leading_lepton_idx = ak.singletons(ak.argmax(lepton.pt, axis=1))
leading_lepton = lepton[leading_lepton_idx]
trailing_lepton_idx = ak.singletons(ak.argmin(lepton.pt, axis=1))
trailing_lepton = lepton[trailing_lepton_idx]
n_nonprompt = getNonPromptFromFlavour(electron) + getNonPromptFromFlavour(muon)
n_chargeflip = getChargeFlips(electron, ev.GenPart) + getChargeFlips(muon, ev.GenPart)
## Jets
jet = getJets(ev, minPt=25, maxEta=4.7, pt_var='pt_nom')
jet = jet[ak.argsort(jet.pt_nom, ascending=False)] # need to sort wrt smeared and recorrected jet pt
jet = jet[~match(jet, muon, deltaRCut=0.4)] # remove jets that overlap with muons
jet = jet[~match(jet, electron, deltaRCut=0.4)] # remove jets that overlap with electrons
central = jet[(abs(jet.eta)<2.4)]
btag = getBTagsDeepFlavB(jet, year=self.year) # should study working point for DeepJet
light = getBTagsDeepFlavB(jet, year=self.year, invert=True)
fwd = getFwdJet(light)
fwd_noPU = getFwdJet(light, puId=False)
## forward jets
j_fwd = fwd[ak.singletons(ak.argmax(fwd.p, axis=1))] # highest momentum spectator
jf = cross(j_fwd, jet)
mjf = (jf['0']+jf['1']).mass
j_fwd2 = jf[ak.singletons(ak.argmax(mjf, axis=1))]['1'] # this is the jet that forms the largest invariant mass with j_fwd
delta_eta = abs(j_fwd2.eta - j_fwd.eta)
## MET -> can switch to puppi MET
met_pt = ev.MET.pt
met_phi = ev.MET.phi
## other variables
ht = ak.sum(jet.pt, axis=1)
st = met_pt + ht + ak.sum(muon.pt, axis=1) + ak.sum(electron.pt, axis=1)
## event selectors
filters = getFilters(ev, year=self.year, dataset=dataset)
dilep = ((ak.num(electron) + ak.num(muon))==2)
pos_charge = ((ak.sum(electron.pdgId, axis=1) + ak.sum(muon.pdgId, axis=1))<0)
neg_charge = ((ak.sum(electron.pdgId, axis=1) + ak.sum(muon.pdgId, axis=1))>0)
lep0pt = ((ak.num(electron[(electron.pt>25)]) + ak.num(muon[(muon.pt>25)]))>0)
lep0pt_40 = ((ak.num(electron[(electron.pt>40)]) + ak.num(muon[(muon.pt>40)]))>0)
lep0pt_100 = ((ak.num(electron[(electron.pt>100)]) + ak.num(muon[(muon.pt>100)]))>0)
lep1pt = ((ak.num(electron[(electron.pt>20)]) + ak.num(muon[(muon.pt>20)]))>1)
lep1pt_30 = ((ak.num(electron[(electron.pt>30)]) + ak.num(muon[(muon.pt>30)]))>1)
lepveto = ((ak.num(vetoelectron) + ak.num(vetomuon))==2)
# define the weight
weight = Weights( len(ev) )
#mult = 1
#if dataset=='inclusive': mult = 0.0478/47.448
#if dataset=='plus': mult = 0.0036/7.205
if not dataset=='MuonEG':
# lumi weight
weight.add("weight", ev.weight*cfg['lumi'][self.year])
#weight.add("weight", ev.genWeight*cfg['lumi'][self.year]*mult)
# PU weight - not in the babies...
weight.add("PU", ev.puWeight, weightUp=ev.puWeightUp, weightDown=ev.puWeightDown, shift=False)
# b-tag SFs
weight.add("btag", self.btagSF.Method1a(btag, light))
# lepton SFs
weight.add("lepton", self.leptonSF.get(electron, muon))
selection = PackedSelection()
selection.add('lepveto', lepveto)
selection.add('dilep', dilep )
selection.add('filter', (filters) )
selection.add('p_T(lep0)>25', lep0pt )
selection.add('p_T(lep0)>40', lep0pt_40 )
selection.add('p_T(lep1)>20', lep1pt )
selection.add('p_T(lep1)>30', lep1pt_30 )
selection.add('SS', ( SSlepton | SSelectron | SSmuon) )
selection.add('pos', ( pos_charge ) )
selection.add('neg', ( neg_charge ) )
selection.add('N_jet>3', (ak.num(jet)>=4) )
selection.add('N_jet>4', (ak.num(jet)>=5) )
selection.add('N_central>2', (ak.num(central)>=3) )
selection.add('N_central>3', (ak.num(central)>=4) )
selection.add('N_btag>0', (ak.num(btag)>=1) )
selection.add('MET>50', (ev.MET.pt>50) )
selection.add('ST', (st>600) )
selection.add('N_fwd>0', (ak.num(fwd)>=1 ))
selection.add('delta_eta', (ak.any(delta_eta>2, axis=1) ) )
selection.add('fwd_p>500', (ak.any(j_fwd.p>500, axis=1) ) )
ss_reqs = ['lepveto', 'dilep', 'SS', 'filter', 'p_T(lep0)>25', 'p_T(lep1)>20', 'N_jet>3', 'N_central>2', 'N_btag>0']
bl_reqs = ss_reqs + ['N_fwd>0', 'N_jet>4', 'N_central>3', 'ST', 'MET>50', 'delta_eta']
sr_reqs = bl_reqs + ['fwd_p>500', 'p_T(lep0)>40', 'p_T(lep1)>30']
ss_reqs_d = { sel: True for sel in ss_reqs }
ss_selection = selection.require(**ss_reqs_d)
bl_reqs_d = { sel: True for sel in bl_reqs }
BL = selection.require(**bl_reqs_d)
sr_reqs_d = { sel: True for sel in sr_reqs }
SR = selection.require(**sr_reqs_d)
cutflow = Cutflow(output, ev, weight=weight)
cutflow_reqs_d = {}
for req in sr_reqs:
cutflow_reqs_d.update({req: True})
cutflow.addRow( req, selection.require(**cutflow_reqs_d) )
# first, make a few super inclusive plots
output['PV_npvs'].fill(dataset=dataset, multiplicity=ev.PV[ss_selection].npvs, weight=weight.weight()[ss_selection])
output['PV_npvsGood'].fill(dataset=dataset, multiplicity=ev.PV[ss_selection].npvsGood, weight=weight.weight()[ss_selection])
output['N_jet'].fill(dataset=dataset, multiplicity=ak.num(jet)[ss_selection], weight=weight.weight()[ss_selection])
output['N_b'].fill(dataset=dataset, multiplicity=ak.num(btag)[ss_selection], weight=weight.weight()[ss_selection])
output['N_central'].fill(dataset=dataset, multiplicity=ak.num(central)[ss_selection], weight=weight.weight()[ss_selection])
output['N_ele'].fill(dataset=dataset, multiplicity=ak.num(electron)[ss_selection], weight=weight.weight()[ss_selection])
output['N_mu'].fill(dataset=dataset, multiplicity=ak.num(electron)[ss_selection], weight=weight.weight()[ss_selection])
output['N_fwd'].fill(dataset=dataset, multiplicity=ak.num(fwd)[ss_selection], weight=weight.weight()[ss_selection])
output['nLepFromTop'].fill(dataset=dataset, multiplicity=ev[BL].nLepFromTop, weight=weight.weight()[BL])
output['nLepFromTau'].fill(dataset=dataset, multiplicity=ev.nLepFromTau[BL], weight=weight.weight()[BL])
output['nLepFromZ'].fill(dataset=dataset, multiplicity=ev.nLepFromZ[BL], weight=weight.weight()[BL])
output['nLepFromW'].fill(dataset=dataset, multiplicity=ev.nLepFromW[BL], weight=weight.weight()[BL])
output['nGenTau'].fill(dataset=dataset, multiplicity=ev.nGenTau[BL], weight=weight.weight()[BL])
output['nGenL'].fill(dataset=dataset, multiplicity=ak.num(ev.GenL[BL], axis=1), weight=weight.weight()[BL])
output['chargeFlip_vs_nonprompt'].fill(dataset=dataset, n1=n_chargeflip[ss_selection], n2=n_nonprompt[ss_selection], n_ele=ak.num(electron)[ss_selection], weight=weight.weight()[ss_selection])
output['MET'].fill(
dataset = dataset,
pt = ev.MET[ss_selection].pt,
phi = ev.MET[ss_selection].phi,
weight = weight.weight()[ss_selection]
)
output['lead_gen_lep'].fill(
dataset = dataset,
pt = ak.to_numpy(ak.flatten(leading_gen_lep[BL].pt)),
eta = ak.to_numpy(ak.flatten(leading_gen_lep[BL].eta)),
phi = ak.to_numpy(ak.flatten(leading_gen_lep[BL].phi)),
weight = weight.weight()[BL]
)
output['trail_gen_lep'].fill(
dataset = dataset,
pt = ak.to_numpy(ak.flatten(trailing_gen_lep[BL].pt)),
eta = ak.to_numpy(ak.flatten(trailing_gen_lep[BL].eta)),
phi = ak.to_numpy(ak.flatten(trailing_gen_lep[BL].phi)),
weight = weight.weight()[BL]
)
output['lead_lep'].fill(
dataset = dataset,
pt = ak.to_numpy(ak.flatten(leading_lepton[BL].pt)),
eta = ak.to_numpy(ak.flatten(leading_lepton[BL].eta)),
phi = ak.to_numpy(ak.flatten(leading_lepton[BL].phi)),
weight = weight.weight()[BL]
)
output['trail_lep'].fill(
dataset = dataset,
pt = ak.to_numpy(ak.flatten(trailing_lepton[BL].pt)),
eta = ak.to_numpy(ak.flatten(trailing_lepton[BL].eta)),
phi = ak.to_numpy(ak.flatten(trailing_lepton[BL].phi)),
weight = weight.weight()[BL]
)
output['j1'].fill(
dataset = dataset,
pt = ak.flatten(jet.pt_nom[:, 0:1][BL]),
eta = ak.flatten(jet.eta[:, 0:1][BL]),
phi = ak.flatten(jet.phi[:, 0:1][BL]),
weight = weight.weight()[BL]
)
output['j2'].fill(
dataset = dataset,
pt = ak.flatten(jet[:, 1:2][BL].pt_nom),
eta = ak.flatten(jet[:, 1:2][BL].eta),
phi = ak.flatten(jet[:, 1:2][BL].phi),
weight = weight.weight()[BL]
)
output['j3'].fill(
dataset = dataset,
pt = ak.flatten(jet[:, 2:3][BL].pt_nom),
eta = ak.flatten(jet[:, 2:3][BL].eta),
phi = ak.flatten(jet[:, 2:3][BL].phi),
weight = weight.weight()[BL]
)
return output
def process(self, events):
output = self._accumulator.identity()
dataset_name = events.metadata['dataset']
output["total_events"][dataset_name] += events.__len__()
# Initialize dict accumulators, if have not been initialized
for jet in [0, 1, 2]:
if dataset_name not in output[f"eta_{jet}_final"].keys():
output[f"eta_{jet}_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"ptoverM_{jet}_final"].keys():
output[f"ptoverM_{jet}_final"][dataset_name] = processor.column_accumulator(np.array([]))
for pair in [(0, 1), (1, 2), (2, 0)]:
if dataset_name not in output[f"dEta_{pair[0]}{pair[1]}_final"].keys():
output[f"dEta_{pair[0]}{pair[1]}_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dR_{pair[0]}{pair[1]}_final"].keys():
output[f"dR_{pair[0]}{pair[1]}_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"moverM_{pair[0]}{pair[1]}_final"].keys():
output[f"moverM_{pair[0]}{pair[1]}_final"][dataset_name] = processor.column_accumulator(np.array([]))
for pair in [(0, 1, 2), (1, 2, 0), (2, 0, 1)]:
if dataset_name not in output[f"dR_{pair[0]}_{pair[1]}{pair[2]}_final"].keys():
output[f"dR_{pair[0]}_{pair[1]}{pair[2]}_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dEta_{pair[0]}_{pair[1]}{pair[2]}_final"].keys():
output[f"dEta_{pair[0]}_{pair[1]}{pair[2]}_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"Phi_{pair[0]}_{pair[1]}{pair[2]}_final"].keys():
output[f"Phi_{pair[0]}_{pair[1]}{pair[2]}_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dPtoverM_{pair[0]}_{pair[1]}{pair[2]}_final"].keys():
output[f"dPtoverM_{pair[0]}_{pair[1]}{pair[2]}_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"ptoverM_max_final"].keys():
output[f"ptoverM_max_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"ptoverM_min_final"].keys():
output[f"ptoverM_min_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"eta_max_final"].keys():
output[f"eta_max_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dR_max_final"].keys():
output[f"dR_max_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dR_min_final"].keys():
output[f"dR_min_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dEta_max_final"].keys():
output[f"dEta_max_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dEta_min_final"].keys():
output[f"dEta_min_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dR_j_jj_max_final"].keys():
output[f"dR_j_jj_max_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dR_j_jj_min_final"].keys():
output[f"dR_j_jj_min_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dEta_j_jj_max_final"].keys():
output[f"dEta_j_jj_max_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dEta_j_jj_min_final"].keys():
output[f"dEta_j_jj_min_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dPhi_j_jj_max_final"].keys():
output[f"dPhi_j_jj_max_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dPhi_j_jj_min_final"].keys():
output[f"dPhi_j_jj_min_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dPtoverM_j_jj_max_final"].keys():
output[f"dPtoverM_j_jj_max_final"][dataset_name] = processor.column_accumulator(np.array([]))
if dataset_name not in output[f"dPtoverM_j_jj_min_final"].keys():
output[f"dPtoverM_j_jj_min_final"][dataset_name] = processor.column_accumulator(np.array([]))
# HLT selection
HLT_mask = []
if year == "2016":
if "SingleMuon" in dataset_name: #this does not work, as the name of file which is under processing is unknown
if "2016B2" in dataset_name:
HLT_mask = events.HLT.IsoMu24 | events.HLT.IsoTkMu24 | events.HLT.Mu50
else:
HLT_mask = events.HLT.IsoMu24 | events.HLT.IsoTkMu24 | events.HLT.Mu50 | events.HLT.TkMu50
else: #https://twiki.cern.ch/twiki/bin/view/CMS/HLTPathsRunIIList
if "2016B2" in dataset_name:
HLT_mask = events.HLT.PFHT800 | events.HLT.PFHT900 | events.HLT.PFJet500 | events.HLT.CaloJet500_NoJetID
elif "2016H" in dataset_name:
HLT_mask = events.HLT.PFHT900 | events.HLT.AK8PFJet450 | events.HLT.AK8PFJet500 | events.HLT.PFJet500 | events.HLT.CaloJet500_NoJetID
else:
HLT_mask = events.HLT.PFHT800 | events.HLT.PFHT900 | events.HLT.AK8PFJet450 | events.HLT.AK8PFJet500 | events.HLT.PFJet500 | events.HLT.CaloJet500_NoJetID
if year == "2017":
if "SingleMuon" in dataset_name:
if "2017B" in dataset_name:
HLT_mask = events.HLT.IsoMu27 | events.HLT.Mu50
else:
HLT_mask = events.HLT.IsoMu27 | events.HLT.Mu50 | events.HLT.OldMu100 | events.HLT.TkMu100
else:
HLT_mask = events.HLT.PFHT1050 | events.HLT.AK8PFJet500 | events.HLT.AK8PFJet550 | events.HLT.CaloJet500_NoJetID | events.HLT.CaloJet550_NoJetID | events.HLT.PFJet500
if year == "2018":
if "SingleMuon" in dataset_name:
HLT_mask = events.HLT.IsoMu24 | events.HLT.Mu50 | events.HLT.OldMu100 | events.HLT.TkMu100
else:
HLT_mask = events.HLT.PFHT1050 | events.HLT.AK8PFJet500 | events.HLT.AK8PFJet550 | events.HLT.CaloJet500_NoJetID | events.HLT.CaloJet550_NoJetID | events.HLT.PFJet500
# Require 3 jets
jet_mask = (events.Jet.pt > 30.) & (abs(events.Jet.eta) < 2.5) & (events.Jet.isTight)
event_mask = (awk.sum(jet_mask, axis=1) >= 3)
event_mask = event_mask & HLT_mask
events_3j = events[event_mask]
# Reduce jet mask to only events with 3 good jets
jet_mask = jet_mask[event_mask]
# Array of the jets to consider for trijet resonance
selected_jets = events_3j.Jet[jet_mask][:, :3]
# Pairs of jets
#pairs = awk.argcombinations(selected_jets, 2)
#jet_i, jet_j = awk.unzip(pairs)
pairs = [(0, 1), (1, 2), (2, 0)]
jet_i, jet_j = zip(*pairs) # Returns [0, 1, 2] , [1, 2, 0]
m_ij = (selected_jets[:, jet_i] + selected_jets[:, jet_j]).mass
dR_ij = selected_jets[:, jet_i].delta_r(selected_jets[:, jet_j])
dEta_ij = abs(selected_jets[:, jet_i].eta - selected_jets[:, jet_j].eta)
jet_k = [2, 0, 1]
dR_i_jk = selected_jets[:, jet_i].delta_r(selected_jets[:, jet_j] + selected_jets[:, jet_k])
dEta_i_jk = abs(selected_jets[:, jet_i].eta - (selected_jets[:, jet_j] + selected_jets[:, jet_k]).eta)
dPhi_i_jk = abs(selected_jets[:, jet_i].phi - (selected_jets[:, jet_j] + selected_jets[:, jet_k]).phi)
m3j = selected_jets.sum().mass
pt_i_overM = selected_jets.pt / m3j
m_01_overM = m_ij[:,0] / m3j
m_12_overM = m_ij[:,1] / m3j
m_20_overM = m_ij[:,2] / m3j
dPtoverM_0_12 = abs(selected_jets[:, 0].pt - (selected_jets[:, 1] + selected_jets[:, 2]).pt) / m3j
dPtoverM_1_20 = abs(selected_jets[:, 1].pt - (selected_jets[:, 2] + selected_jets[:, 0]).pt) / m3j
dPtoverM_2_01 = abs(selected_jets[:, 2].pt - (selected_jets[:, 0] + selected_jets[:, 1]).pt) / m3j
# Event selection masks
# selection_masks = {}
# Pre-selection
selection = PackedSelection()
selection.add("Dummy", m3j > 000)
sel_mask = selection.require(**{name: True for name in selection.names})
# selection_masks["Pre-selection"] = sel_mask
output["selected_events"][dataset_name] += events_3j[sel_mask].__len__()
for jet in [0, 1, 2]:
output[f"eta_{jet}_final"][dataset_name] += processor.column_accumulator(np.array(selected_jets[:, jet][sel_mask].eta))
output[f"ptoverM_{jet}_final"][dataset_name] += processor.column_accumulator(np.array(pt_i_overM[:, jet][sel_mask]))
for pair in [(0, 1), (1, 2), (2, 0)]:
output[f"dEta_{pair[0]}{pair[1]}_final"][dataset_name] += processor.column_accumulator(np.array(dEta_ij[:, pair[0]][sel_mask]))
output[f"dR_{pair[0]}{pair[1]}_final"][dataset_name] += processor.column_accumulator(np.array(dR_ij[:, pair[0]][sel_mask]))
output[f"moverM_01_final"][dataset_name] += processor.column_accumulator(np.array(m_01_overM[sel_mask]))
output[f"moverM_12_final"][dataset_name] += processor.column_accumulator(np.array(m_12_overM[sel_mask]))
output[f"moverM_20_final"][dataset_name] += processor.column_accumulator(np.array(m_20_overM[sel_mask]))
for pair in [(0, 1, 2), (1, 2, 0), (2, 0, 1)]:
output[f"dR_{pair[0]}_{pair[1]}{pair[2]}_final"][dataset_name] += processor.column_accumulator(np.array(dR_i_jk[:, pair[0]][sel_mask]))
output[f"dEta_{pair[0]}_{pair[1]}{pair[2]}_final"][dataset_name] += processor.column_accumulator(np.array(dEta_i_jk[:, pair[0]][sel_mask]))
output[f"Phi_{pair[0]}_{pair[1]}{pair[2]}_final"][dataset_name] += processor.column_accumulator(np.array(dPhi_i_jk[:, pair[0]][sel_mask]))
output[f"dPtoverM_0_12_final"][dataset_name] += processor.column_accumulator(np.array(dPtoverM_0_12[sel_mask]))
output[f"dPtoverM_1_20_final"][dataset_name] += processor.column_accumulator(np.array(dPtoverM_1_20[sel_mask]))
output[f"dPtoverM_2_01_final"][dataset_name] += processor.column_accumulator(np.array(dPtoverM_2_01[sel_mask]))
max_pt_overM_2fill = awk.max(pt_i_overM[sel_mask], axis=1)
min_pt_overM_2fill = awk.min(pt_i_overM[sel_mask], axis=1)
max_dR_2fill = awk.max(dR_ij[sel_mask], axis=1)
max_dEta_2fill = awk.max(dEta_ij[sel_mask], axis=1)
min_dR_2fill = awk.min(dR_ij[sel_mask], axis=1)
min_dEta_2fill = awk.min(dEta_ij[sel_mask], axis=1)
min_pt_2fill = awk.min(selected_jets[sel_mask].pt, axis=1)
max_eta_2fill = awk.max(abs(selected_jets[sel_mask].eta), axis=1)
max_dR_i_jk_2fill = awk.max(dR_i_jk[sel_mask], axis=1)
min_dR_i_jk_2fill = awk.min(dR_i_jk[sel_mask], axis=1)
max_dEta_i_jk_2fill = awk.max(dEta_i_jk[sel_mask], axis=1)
min_dEta_i_jk_2fill = awk.min(dEta_i_jk[sel_mask], axis=1)
max_dPhi_i_jk_2fill = awk.max(dPhi_i_jk[sel_mask], axis=1)
min_dPhi_i_jk_2fill = awk.min(dPhi_i_jk[sel_mask], axis=1)
max_dPtoverM_i_jk_2fill = []
min_dPtoverM_i_jk_2fill = []
dPtoverM_0_12_2fill = dPtoverM_0_12[sel_mask]
dPtoverM_1_20_2fill = dPtoverM_1_20[sel_mask]
dPtoverM_2_01_2fill = dPtoverM_2_01[sel_mask]
for pair in zip(dPtoverM_0_12_2fill, dPtoverM_1_20_2fill, dPtoverM_2_01_2fill):
max_dPtoverM_i_jk_2fill.append(max(pair))
min_dPtoverM_i_jk_2fill.append(min(pair))
max_pt_overM_2fill = awk.fill_none(max_pt_overM_2fill, -99)
min_pt_overM_2fill = awk.fill_none(min_pt_overM_2fill, -99)
max_dR_2fill = awk.fill_none(max_dR_2fill, -99)
max_dEta_2fill = awk.fill_none(max_dEta_2fill, -99)
min_dR_2fill = awk.fill_none(min_dR_2fill, -99)
min_dEta_2fill = awk.fill_none(min_dEta_2fill, -99)
min_pt_2fill = awk.fill_none(min_pt_2fill, -99)
max_eta_2fill = awk.fill_none(max_eta_2fill, -99)
max_dR_i_jk_2fill = awk.fill_none(max_dR_i_jk_2fill, -99)
min_dR_i_jk_2fill = awk.fill_none(min_dR_i_jk_2fill, -99)
max_dEta_i_jk_2fill = awk.fill_none(max_dEta_i_jk_2fill, -99)
min_dEta_i_jk_2fill = awk.fill_none(min_dEta_i_jk_2fill, -99)
max_dPhi_i_jk_2fill = awk.fill_none(max_dPhi_i_jk_2fill, -99)
min_dPhi_i_jk_2fill = awk.fill_none(min_dPhi_i_jk_2fill, -99)
output[f"ptoverM_max_final"][dataset_name] += processor.column_accumulator(np.array(max_pt_overM_2fill))
output[f"ptoverM_min_final"][dataset_name] += processor.column_accumulator(np.array(min_pt_overM_2fill))
output[f"eta_max_final"][dataset_name] += processor.column_accumulator(np.array(max_eta_2fill))
output[f"dR_max_final"][dataset_name] += processor.column_accumulator(np.array(max_dR_2fill))
output[f"dR_min_final"][dataset_name] += processor.column_accumulator(np.array(min_dR_2fill))
output[f"dEta_max_final"][dataset_name] += processor.column_accumulator(np.array(max_dEta_2fill))
output[f"dEta_min_final"][dataset_name] += processor.column_accumulator(np.array(min_dEta_2fill))
output[f"dR_j_jj_max_final"][dataset_name] += processor.column_accumulator(np.array(max_dR_i_jk_2fill))
output[f"dR_j_jj_min_final"][dataset_name] += processor.column_accumulator(np.array(min_dR_i_jk_2fill))
output[f"dEta_j_jj_max_final"][dataset_name] += processor.column_accumulator(np.array(max_dEta_i_jk_2fill))
output[f"dEta_j_jj_min_final"][dataset_name] += processor.column_accumulator(np.array(min_dEta_i_jk_2fill))
output[f"dPhi_j_jj_max_final"][dataset_name] += processor.column_accumulator(np.array(max_dPhi_i_jk_2fill))
output[f"dPhi_j_jj_min_final"][dataset_name] += processor.column_accumulator(np.array(min_dPhi_i_jk_2fill))
output[f"dPtoverM_j_jj_max_final"][dataset_name] += processor.column_accumulator(np.array(max_dPtoverM_i_jk_2fill))
output[f"dPtoverM_j_jj_min_final"][dataset_name] += processor.column_accumulator(np.array(min_dPtoverM_i_jk_2fill))
return output
def process(self, events):
output = self.accumulator.identity()
# we can use a very loose preselection to filter the events. nothing is done with this presel, though
presel = ak.num(events.Jet) > 0
ev = events[presel]
dataset = ev.metadata['dataset']
# load the config - probably not needed anymore
cfg = loadConfig()
output['totalEvents']['all'] += len(events)
output['skimmedEvents']['all'] += len(ev)
## Electrons
electron = Collections(ev, "Electron", "tightFCNC", 0, self.year).get()
electron = electron[(electron.pt > 15) & (np.abs(electron.eta) < 2.4)]
electron = electron[(electron.genPartIdx >= 0)]
electron = electron[(np.abs(electron.matched_gen.pdgId) == 11
)] #from here on all leptons are gen-matched
electron = electron[((electron.genPartFlav == 1) |
(electron.genPartFlav
== 15))] #and now they are all prompt
is_flipped = (((electron.matched_gen.pdgId * (-1) == electron.pdgId) |
(find_first_parent(electron.matched_gen) *
(-1) == electron.pdgId)) &
(np.abs(electron.pdgId) == 11))
flipped_electron = electron[is_flipped]
n_flips = ak.num(flipped_electron)
leading_electron_idx = ak.singletons(ak.argmax(electron.pt, axis=1))
leading_electron = electron[leading_electron_idx]
leading_flipped_electron_idx = ak.singletons(
ak.argmax(flipped_electron.pt, axis=1))
leading_flipped_electron = electron[leading_flipped_electron_idx]
def getMVAscore(electron):
if self.year == 2016:
MVA = electron.mvaSpring16GP
return MVA
elif self.year == 2017:
MVA = electron.mvaFall17V2noIso
return MVA
elif self.year == 2018:
MVA = np.minimum(
np.maximum(electron.mvaFall17V2noIso, -1.0 + 1.e-6),
1.0 - 1.e-6)
return -0.5 * np.log(2 / (MVA + 1) - 1)
# setting up the various weights
#weight = Weights( len(ev) )
#if not dataset=='MuonEG':
# generator weight
# weight.add("weight", ev.genWeight)
#selections
filters = getFilters(ev, year=self.year, dataset=dataset)
electr = ((ak.num(electron) >= 1))
flip = (n_flips >= 1)
selection = PackedSelection()
selection.add('filter', (filters))
selection.add('electr', electr)
selection.add('flip', flip)
bl_reqs = ['filter', 'electr']
bl_reqs_d = {sel: True for sel in bl_reqs}
baseline = selection.require(**bl_reqs_d)
f_reqs = bl_reqs + ['flip']
f_reqs_d = {sel: True for sel in f_reqs}
flip_sel = selection.require(**f_reqs_d)
#adjust weights to prevent length mismatch
#ak_weight_gen = ak.ones_like(electron[baseline].pt) * weight.weight()[baseline]
#ak_weight_flip = ak.ones_like(flipped_electron[flip_sel].pt) * weight.weight()[flip_sel]
#output['N_ele'].fill(dataset=dataset, multiplicity=ak.num(electron)[baseline], weight=weight.weight()[baseline])
#output['electron_flips'].fill(dataset=dataset, multiplicity=n_flips[baseline], weight=weight.weight()[baseline])
output["electron"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(electron[baseline].pt)),
eta=abs(ak.to_numpy(ak.flatten(electron[baseline].eta))),
)
output["electron2"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(electron[baseline].pt)),
eta=ak.to_numpy(ak.flatten(electron[baseline].eta)),
)
output["flipped_electron"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(flipped_electron[flip_sel].pt)),
eta=abs(ak.to_numpy(ak.flatten(flipped_electron[flip_sel].eta))),
)
output["flipped_electron2"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(flipped_electron[flip_sel].pt)),
eta=ak.to_numpy(ak.flatten(flipped_electron[flip_sel].eta)),
)
output["mva_id"].fill(
dataset=dataset,
mva_id=ak.to_numpy(ak.flatten(getMVAscore(electron)[baseline])),
eta=np.abs(ak.to_numpy(ak.flatten(electron.etaSC[baseline]))),
)
output["mva_id2"].fill(
dataset=dataset,
mva_id=ak.to_numpy(ak.flatten(getMVAscore(electron)[baseline])),
pt=ak.to_numpy(ak.flatten(electron.pt[baseline])),
)
output["isolation"].fill(
dataset=dataset,
isolation1=ak.to_numpy(ak.flatten(electron.jetRelIso[baseline])),
isolation2=ak.to_numpy(ak.flatten(electron.jetPtRelv2[baseline])),
)
return output
def process(self, events):
output = self._accumulator.identity()
dataset_name = events.metadata['dataset']
output["total_events"][dataset_name] += events.__len__()
# HLT selection
HLT_mask = []
if year == "2016":
if "SingleMuon" in dataset_name: #this does not work, as the name of file which is under processing is unknown
if "2016B2" in dataset_name:
HLT_mask = events.HLT.IsoMu24 | events.HLT.IsoTkMu24 | events.HLT.Mu50
else:
HLT_mask = events.HLT.IsoMu24 | events.HLT.IsoTkMu24 | events.HLT.Mu50 | events.HLT.TkMu50
else: #https://twiki.cern.ch/twiki/bin/view/CMS/HLTPathsRunIIList
if "2016B2" in dataset_name:
HLT_mask = events.HLT.PFHT800 | events.HLT.PFHT900 | events.HLT.PFJet500 | events.HLT.CaloJet500_NoJetID
elif "2016H" in dataset_name:
HLT_mask = events.HLT.PFHT900 | events.HLT.AK8PFJet450 | events.HLT.AK8PFJet500 | events.HLT.PFJet500 | events.HLT.CaloJet500_NoJetID
else:
HLT_mask = events.HLT.PFHT800 | events.HLT.PFHT900 | events.HLT.AK8PFJet450 | events.HLT.AK8PFJet500 | events.HLT.PFJet500 | events.HLT.CaloJet500_NoJetID
if year == "2017":
if "SingleMuon" in dataset_name:
if "2017B" in dataset_name:
HLT_mask = events.HLT.IsoMu27 | events.HLT.Mu50
else:
HLT_mask = events.HLT.IsoMu27 | events.HLT.Mu50 | events.HLT.OldMu100 | events.HLT.TkMu100
else:
HLT_mask = events.HLT.PFHT1050 | events.HLT.AK8PFJet500 | events.HLT.AK8PFJet550 | events.HLT.CaloJet500_NoJetID | events.HLT.CaloJet550_NoJetID | events.HLT.PFJet500
if year == "2018":
if "SingleMuon" in dataset_name:
HLT_mask = events.HLT.IsoMu24 | events.HLT.Mu50 | events.HLT.OldMu100 | events.HLT.TkMu100
else:
HLT_mask = events.HLT.PFHT1050 | events.HLT.AK8PFJet500 | events.HLT.AK8PFJet550 | events.HLT.CaloJet500_NoJetID | events.HLT.CaloJet550_NoJetID | events.HLT.PFJet500
# Require 3 jets
jet_mask = (events.Jet.pt > 30.) & (abs(events.Jet.eta) <
2.5) & (events.Jet.isTight)
event_mask = (awk.sum(jet_mask, axis=1) >= 3)
event_mask = event_mask & HLT_mask
events_3j = events[event_mask]
# Reduce jet mask to only events with 3 good jets
jet_mask = jet_mask[event_mask]
# Array of the jets to consider for trijet resonance
selected_jets = events_3j.Jet[jet_mask][:, :3]
# Pairs of jets
#pairs = awk.argcombinations(selected_jets, 2)
#jet_i, jet_j = awk.unzip(pairs)
pairs = [(0, 1), (1, 2), (2, 0)]
jet_i, jet_j = zip(*pairs) # Returns [0, 1, 2] , [1, 2, 0]
m_ij = (selected_jets[:, jet_i] + selected_jets[:, jet_j]).mass
dR_ij = selected_jets[:, jet_i].delta_r(selected_jets[:, jet_j])
dEta_ij = abs(selected_jets[:, jet_i].eta -
selected_jets[:, jet_j].eta)
jet_k = [2, 0, 1]
dR_i_jk = selected_jets[:, jet_i].delta_r(selected_jets[:, jet_j] +
selected_jets[:, jet_k])
dEta_i_jk = abs(selected_jets[:, jet_i].eta -
(selected_jets[:, jet_j] +
selected_jets[:, jet_k]).eta)
dPhi_i_jk = abs(selected_jets[:, jet_i].phi -
(selected_jets[:, jet_j] +
selected_jets[:, jet_k]).phi)
m3j = selected_jets.sum().mass
pt_i_overM = selected_jets.pt / m3j
m_01_overM = m_ij[:, 0] / m3j
m_12_overM = m_ij[:, 1] / m3j
m_20_overM = m_ij[:, 2] / m3j
dPtoverM_0_12 = abs(selected_jets[:, 0].pt -
(selected_jets[:, 1] +
selected_jets[:, 2]).pt) / m3j
dPtoverM_1_20 = abs(selected_jets[:, 1].pt -
(selected_jets[:, 2] +
selected_jets[:, 0]).pt) / m3j
dPtoverM_2_01 = abs(selected_jets[:, 2].pt -
(selected_jets[:, 0] +
selected_jets[:, 1]).pt) / m3j
# Event selection masks
selection_masks = {}
# Pre-selection
selection = PackedSelection()
selection.add("Dummy", m3j > 000)
sel_mask = selection.require(
**{name: True
for name in selection.names})
selection_masks["Pre-selection"] = sel_mask
# HLT_trigger (this is already done at the beginning)
# if year == "2016":
# JetHLT_mask = []
# if "2016B2" in dataset_name:
# JetHLT_mask = events.HLT.PFHT800 | events.HLT.PFHT900 | events.HLT.PFJet500 | events.HLT.CaloJet500_NoJetID
# elif "2016H" in dataset_name:
# JetHLT_mask = events.HLT.PFHT900 | events.HLT.AK8PFJet450 | events.HLT.AK8PFJet500 | events.HLT.PFJet500 | events.HLT.CaloJet500_NoJetID
# else:
# JetHLT_mask = events.HLT.PFHT800 | events.HLT.PFHT900 | events.HLT.AK8PFJet450 | events.HLT.AK8PFJet500 | events.HLT.PFJet500 | events.HLT.CaloJet500_NoJetID
# selection_masks["JetHLT"] = JetHLT_mask[event_mask]
# if year == "2017":
# JetHLT_mask = events.HLT.PFHT1050 | events.HLT.AK8PFJet500 | events.HLT.AK8PFJet550 | events.HLT.CaloJet500_NoJetID | events.HLT.CaloJet550_NoJetID | events.HLT.PFJet500
# selection_masks["JetHLT"] = JetHLT_mask[event_mask]
# if year == "2018":
# JetHLT_mask = events.HLT.PFHT1050 | events.HLT.AK8PFJet500 | events.HLT.AK8PFJet550 | events.HLT.CaloJet500_NoJetID | events.HLT.CaloJet550_NoJetID | events.HLT.PFJet500
# selection_masks["JetHLT"] = JetHLT_mask[event_mask]
# Fill histograms
for selection, selection_mask in selection_masks.items():
output["mjjj"].fill(dataset=dataset_name,
selection=selection,
mjjj=m3j[selection_mask])
output["m_ij"].fill(dataset=dataset_name,
selection=selection,
m_01=m_ij[:, 0][selection_mask],
m_12=m_ij[:, 1][selection_mask],
m_20=m_ij[:, 2][selection_mask])
output["dR_ij"].fill(dataset=dataset_name,
selection=selection,
dR_01=dR_ij[:, 0][selection_mask],
dR_12=dR_ij[:, 1][selection_mask],
dR_20=dR_ij[:, 2][selection_mask])
output["dEta_ij"].fill(dataset=dataset_name,
selection=selection,
dEta_01=dEta_ij[:, 0][selection_mask],
dEta_12=dEta_ij[:, 1][selection_mask],
dEta_20=dEta_ij[:, 2][selection_mask])
output["moverM_ij"].fill(dataset=dataset_name,
selection=selection,
moverM_01=m_01_overM[selection_mask],
moverM_12=m_12_overM[selection_mask],
moverM_20=m_20_overM[selection_mask])
output["pt_i"].fill(dataset=dataset_name,
selection=selection,
pt_0=selected_jets[:, 0][selection_mask].pt,
pt_1=selected_jets[:, 1][selection_mask].pt,
pt_2=selected_jets[:, 2][selection_mask].pt)
output["eta_i"].fill(dataset=dataset_name,
selection=selection,
eta_0=selected_jets[:, 0][selection_mask].eta,
eta_1=selected_jets[:, 1][selection_mask].eta,
eta_2=selected_jets[:, 2][selection_mask].eta)
output["ptoverM_i"].fill(dataset=dataset_name,
selection=selection,
ptoverM_0=pt_i_overM[:,
0][selection_mask],
ptoverM_1=pt_i_overM[:,
1][selection_mask],
ptoverM_2=pt_i_overM[:,
2][selection_mask])
output["dR_i_jk"].fill(dataset=dataset_name,
selection=selection,
dR_0_12=dR_i_jk[:, 0][selection_mask],
dR_1_20=dR_i_jk[:, 1][selection_mask],
dR_2_01=dR_i_jk[:, 2][selection_mask])
output["dEta_i_jk"].fill(dataset=dataset_name,
selection=selection,
dEta_0_12=dEta_i_jk[:, 0][selection_mask],
dEta_1_20=dEta_i_jk[:, 1][selection_mask],
dEta_2_01=dEta_i_jk[:, 2][selection_mask])
output["dPhi_i_jk"].fill(dataset=dataset_name,
selection=selection,
dPhi_0_12=dPhi_i_jk[:, 0][selection_mask],
dPhi_1_20=dPhi_i_jk[:, 1][selection_mask],
dPhi_2_01=dPhi_i_jk[:, 2][selection_mask])
output["dPtoverM_i_jk"].fill(
dataset=dataset_name,
selection=selection,
dPtoverM_0_12=dPtoverM_0_12[selection_mask],
dPtoverM_1_20=dPtoverM_1_20[selection_mask],
dPtoverM_2_01=dPtoverM_2_01[selection_mask])
pt_i_overM_2fill = pt_i_overM[selection_mask]
dR_ij_2fill = dR_ij[selection_mask]
dEta_ij_2fill = dEta_ij[selection_mask]
dR_i_jk_2fill = dR_i_jk[selection_mask]
dEta_i_jk_2fill = dEta_i_jk[selection_mask]
dPhi_i_jk_2fill = dPhi_i_jk[selection_mask]
dPtoverM_0_12_2fill = dPtoverM_0_12[selection_mask]
dPtoverM_1_20_2fill = dPtoverM_1_20[selection_mask]
dPtoverM_2_01_2fill = dPtoverM_2_01[selection_mask]
selected_jets_2fill = selected_jets[selection_mask]
max_pt_overM_2fill = awk.max(pt_i_overM_2fill, axis=1)
min_pt_overM_2fill = awk.min(pt_i_overM_2fill, axis=1)
max_dR_2fill = awk.max(dR_ij_2fill, axis=1)
max_dEta_2fill = awk.max(dEta_ij_2fill, axis=1)
min_dR_2fill = awk.min(dR_ij_2fill, axis=1)
min_dEta_2fill = awk.min(dEta_ij_2fill, axis=1)
min_pt_2fill = awk.min(selected_jets_2fill.pt, axis=1)
max_eta_2fill = awk.max(abs(selected_jets_2fill.eta), axis=1)
max_dR_i_jk_2fill = awk.max(dR_i_jk_2fill, axis=1)
min_dR_i_jk_2fill = awk.min(dR_i_jk_2fill, axis=1)
max_dEta_i_jk_2fill = awk.max(dEta_i_jk_2fill, axis=1)
min_dEta_i_jk_2fill = awk.min(dEta_i_jk_2fill, axis=1)
max_dPhi_i_jk_2fill = awk.max(dPhi_i_jk_2fill, axis=1)
min_dPhi_i_jk_2fill = awk.min(dPhi_i_jk_2fill, axis=1)
max_dPtoverM_i_jk_2fill = []
min_dPtoverM_i_jk_2fill = []
for pair in zip(dPtoverM_0_12_2fill, dPtoverM_1_20_2fill,
dPtoverM_2_01_2fill):
max_dPtoverM_i_jk_2fill.append(max(pair))
min_dPtoverM_i_jk_2fill.append(min(pair))
max_dPtoverM_i_jk_2fill = np.array(max_dPtoverM_i_jk_2fill)
min_dPtoverM_i_jk_2fill = np.array(min_dPtoverM_i_jk_2fill)
max_pt_overM_2fill = awk.fill_none(max_pt_overM_2fill, -99)
min_pt_overM_2fill = awk.fill_none(min_pt_overM_2fill, -99)
max_dR_2fill = awk.fill_none(max_dR_2fill, -99)
max_dEta_2fill = awk.fill_none(max_dEta_2fill, -99)
min_dR_2fill = awk.fill_none(min_dR_2fill, -99)
min_dEta_2fill = awk.fill_none(min_dEta_2fill, -99)
min_pt_2fill = awk.fill_none(min_pt_2fill, -99)
max_eta_2fill = awk.fill_none(max_eta_2fill, -99)
max_dR_i_jk_2fill = awk.fill_none(max_dR_i_jk_2fill, -99)
min_dR_i_jk_2fill = awk.fill_none(min_dR_i_jk_2fill, -99)
max_dEta_i_jk_2fill = awk.fill_none(max_dEta_i_jk_2fill, -99)
min_dEta_i_jk_2fill = awk.fill_none(min_dEta_i_jk_2fill, -99)
max_dPhi_i_jk_2fill = awk.fill_none(max_dPhi_i_jk_2fill, -99)
min_dPhi_i_jk_2fill = awk.fill_none(min_dPhi_i_jk_2fill, -99)
output["max_dR"].fill(dataset=dataset_name,
selection=selection,
max_dR=max_dR_2fill)
output["max_dEta"].fill(dataset=dataset_name,
selection=selection,
max_dEta=max_dEta_2fill)
output["min_dR"].fill(dataset=dataset_name,
selection=selection,
min_dR=min_dR_2fill)
output["min_dEta"].fill(dataset=dataset_name,
selection=selection,
min_dEta=min_dEta_2fill)
output["min_pt"].fill(dataset=dataset_name,
selection=selection,
min_pt=min_pt_2fill)
output["max_eta"].fill(dataset=dataset_name,
selection=selection,
max_eta=max_eta_2fill)
output["max_ptoverM"].fill(dataset=dataset_name,
selection=selection,
max_ptoverM=max_pt_overM_2fill)
output["min_ptoverM"].fill(dataset=dataset_name,
selection=selection,
min_ptoverM=min_pt_overM_2fill)
output["max_dR_j_jj"].fill(dataset=dataset_name,
selection=selection,
max_dR_j_jj=max_dR_i_jk_2fill)
output["max_dEta_j_jj"].fill(dataset=dataset_name,
selection=selection,
max_dEta_j_jj=max_dEta_i_jk_2fill)
output["max_dPhi_j_jj"].fill(dataset=dataset_name,
selection=selection,
max_dPhi_j_jj=max_dPhi_i_jk_2fill)
output["max_dPtoverM_j_jj"].fill(
dataset=dataset_name,
selection=selection,
max_dPtoverM_j_jj=max_dPtoverM_i_jk_2fill)
output["min_dR_j_jj"].fill(dataset=dataset_name,
selection=selection,
min_dR_j_jj=min_dR_i_jk_2fill)
output["min_dEta_j_jj"].fill(dataset=dataset_name,
selection=selection,
min_dEta_j_jj=min_dEta_i_jk_2fill)
output["min_dPhi_j_jj"].fill(dataset=dataset_name,
selection=selection,
min_dPhi_j_jj=min_dPhi_i_jk_2fill)
output["min_dPtoverM_j_jj"].fill(
dataset=dataset_name,
selection=selection,
min_dPtoverM_j_jj=min_dPtoverM_i_jk_2fill)
return output
def test_packed_selection():
from coffea.analysis_tools import PackedSelection
sel = PackedSelection()
shape = (10, )
all_true = np.full(shape=shape, fill_value=True, dtype=np.bool)
all_false = np.full(shape=shape, fill_value=False, dtype=np.bool)
fizz = np.arange(shape[0]) % 3 == 0
buzz = np.arange(shape[0]) % 5 == 0
ones = np.ones(shape=shape, dtype=np.uint64)
wrong_shape = ones = np.ones(shape=(shape[0] - 5, ), dtype=np.bool)
sel.add("all_true", all_true)
sel.add("all_false", all_false)
sel.add("fizz", fizz)
sel.add("buzz", buzz)
assert np.all(sel.require(all_true=True, all_false=False) == all_true)
# allow truthy values
assert np.all(sel.require(all_true=1, all_false=0) == all_true)
assert np.all(sel.all("all_true", "all_false") == all_false)
assert np.all(sel.any("all_true", "all_false") == all_true)
assert np.all(
sel.all("fizz", "buzz") == np.array([
True, False, False, False, False, False, False, False, False, False
]))
assert np.all(
sel.any("fizz", "buzz") == np.array(
[True, False, False, True, False, True, True, False, False, True]))
with pytest.raises(ValueError):
sel.add("wrong_shape", wrong_shape)
with pytest.raises(ValueError):
sel.add("ones", ones)
with pytest.raises(RuntimeError):
overpack = PackedSelection()
for i in range(65):
overpack.add("sel_%d", all_true)
class Selection:
def __init__(self, **kwargs):
'''
kwargs should be:
ele (loose and tight)
mu
jets: all, central, forward, b-tag
met
'''
self.__dict__.update(kwargs)
## not yet sure whether this should go here, or later
#self.filters = getFilters(self.events, year=self.year, dataset=self.dataset)
def dilep_baseline(self, omit=[], cutflow=None, tight=False, SS=True):
'''
give it a cutflow object if you want it to be filed.
cuts in the omit list will not be applied
'''
self.selection = PackedSelection()
is_dilep = ((ak.num(self.ele) + ak.num(self.mu))==2)
pos_charge = ((ak.sum(self.ele.pdgId, axis=1) + ak.sum(self.mu.pdgId, axis=1))<0)
neg_charge = ((ak.sum(self.ele.pdgId, axis=1) + ak.sum(self.mu.pdgId, axis=1))>0)
lep0pt = ((ak.num(self.ele[(self.ele.pt>25)]) + ak.num(self.mu[(self.mu.pt>25)]))>0)
lep1pt = ((ak.num(self.ele[(self.ele.pt>20)]) + ak.num(self.mu[(self.mu.pt>20)]))>1)
lepveto = ((ak.num(self.ele_veto) + ak.num(self.mu_veto))==2)
dimu = choose(self.mu, 2)
diele = choose(self.ele, 2)
dilep = cross(self.mu, self.ele)
if SS:
is_SS = ( ak.any((dimu['0'].charge * dimu['1'].charge)>0, axis=1) | \
ak.any((diele['0'].charge * diele['1'].charge)>0, axis=1) | \
ak.any((dilep['0'].charge * dilep['1'].charge)>0, axis=1) )
else:
is_OS = ( ak.any((dimu['0'].charge * dimu['1'].charge)<0, axis=1) | \
ak.any((diele['0'].charge * diele['1'].charge)<0, axis=1) | \
ak.any((dilep['0'].charge * dilep['1'].charge)<0, axis=1) )
lepton = ak.concatenate([self.ele, self.mu], axis=1)
lepton_pdgId_pt_ordered = ak.fill_none(
ak.pad_none(
lepton[ak.argsort(lepton.pt, ascending=False)].pdgId, 2, clip=True),
0)
triggers = getTriggers(self.events,
ak.flatten(lepton_pdgId_pt_ordered[:,0:1]),
ak.flatten(lepton_pdgId_pt_ordered[:,1:2]), year=self.year, dataset=self.dataset)
ht = ak.sum(self.jet_all.pt, axis=1)
st = self.met.pt + ht + ak.sum(self.mu.pt, axis=1) + ak.sum(self.ele.pt, axis=1)
self.selection.add('lepveto', lepveto)
self.selection.add('dilep', is_dilep)
#self.selection.add('filter', self.filters)
self.selection.add('trigger', triggers)
self.selection.add('p_T(lep0)>25', lep0pt)
self.selection.add('p_T(lep1)>20', lep1pt)
if SS:
self.selection.add('SS', is_SS )
else:
self.selection.add('OS', is_OS )
self.selection.add('N_jet>3', (ak.num(self.jet_all)>3) )
self.selection.add('N_jet>4', (ak.num(self.jet_all)>4) )
self.selection.add('N_central>2', (ak.num(self.jet_central)>2) )
self.selection.add('N_central>3', (ak.num(self.jet_central)>3) )
self.selection.add('N_btag>0', (ak.num(self.jet_btag)>0) )
self.selection.add('N_fwd>0', (ak.num(self.jet_fwd)>0) )
self.selection.add('MET>30', (self.met.pt>30) )
self.selection.add('MET>50', (self.met.pt>50) )
self.selection.add('ST>600', (st>600) )
ss_reqs = [
# 'filter',
'lepveto',
'dilep',
'p_T(lep0)>25',
'p_T(lep1)>20',
'trigger',
'SS' if SS else 'OS',
'N_jet>3',
'N_central>2',
'N_btag>0',
'MET>30',
'N_fwd>0',
]
if tight:
ss_reqs += [
'N_jet>4',
'N_central>3',
'ST>600',
'MET>50',
#'delta_eta',
]
ss_reqs_d = { sel: True for sel in ss_reqs if not sel in omit }
ss_selection = self.selection.require(**ss_reqs_d)
if cutflow:
#
cutflow_reqs_d = {}
for req in ss_reqs:
cutflow_reqs_d.update({req: True})
cutflow.addRow( req, self.selection.require(**cutflow_reqs_d) )
return ss_selection
def trilep_baseline(self, omit=[], cutflow=None, tight=False):
'''
give it a cutflow object if you want it to be filed.
cuts in the omit list will not be applied
'''
self.selection = PackedSelection()
is_trilep = ((ak.num(self.ele) + ak.num(self.mu))==3)
pos_charge = ((ak.sum(self.ele.pdgId, axis=1) + ak.sum(self.mu.pdgId, axis=1))<0)
neg_charge = ((ak.sum(self.ele.pdgId, axis=1) + ak.sum(self.mu.pdgId, axis=1))>0)
lep0pt = ((ak.num(self.ele[(self.ele.pt>25)]) + ak.num(self.mu[(self.mu.pt>25)]))>0)
lep1pt = ((ak.num(self.ele[(self.ele.pt>20)]) + ak.num(self.mu[(self.mu.pt>20)]))>1)
lepveto = ((ak.num(self.ele_veto) + ak.num(self.mu_veto))==3)
dimu = choose(self.mu, 2)
diele = choose(self.ele, 2)
dilep = cross(self.mu, self.ele)
OS_dimu = dimu[(dimu['0'].charge*dimu['1'].charge < 0)]
OS_diele = diele[(diele['0'].charge*diele['1'].charge < 0)]
offZ = (ak.all(abs(OS_dimu.mass-91.2)>10, axis=1) & ak.all(abs(OS_diele.mass-91.2)>10, axis=1))
lepton = ak.concatenate([self.ele, self.mu], axis=1)
lepton_pdgId_pt_ordered = ak.fill_none(ak.pad_none(lepton[ak.argsort(lepton.pt, ascending=False)].pdgId, 2, clip=True), 0)
triggers = getTriggers(self.events,
ak.flatten(lepton_pdgId_pt_ordered[:,0:1]),
ak.flatten(lepton_pdgId_pt_ordered[:,1:2]), year=self.year, dataset=self.dataset)
ht = ak.sum(self.jet_all.pt, axis=1)
st = self.met.pt + ht + ak.sum(self.mu.pt, axis=1) + ak.sum(self.ele.pt, axis=1)
self.selection.add('lepveto', lepveto)
self.selection.add('trilep', is_trilep)
self.selection.add('filter', self.filters)
self.selection.add('trigger', triggers)
self.selection.add('p_T(lep0)>25', lep0pt)
self.selection.add('p_T(lep1)>20', lep1pt)
self.selection.add('N_jet>2', (ak.num(self.jet_all)>2) )
self.selection.add('N_jet>3', (ak.num(self.jet_all)>3) )
self.selection.add('N_central>1', (ak.num(self.jet_central)>1) )
self.selection.add('N_central>2', (ak.num(self.jet_central)>2) )
self.selection.add('N_btag>0', (ak.num(self.jet_btag)>0) )
self.selection.add('N_fwd>0', (ak.num(self.jet_fwd)>0) )
self.selection.add('MET>50', (self.met.pt>50) )
self.selection.add('ST>600', (st>600) )
self.selection.add('offZ', offZ )
reqs = [
'filter',
'lepveto',
'trilep',
'p_T(lep0)>25',
'p_T(lep1)>20',
'trigger',
'offZ',
'MET>50',
'N_jet>2',
'N_central>1',
'N_btag>0',
'N_fwd>0',
]
if tight:
reqs += [
'N_jet>3',
'N_central>2',
'ST>600',
#'MET>50',
#'delta_eta',
]
reqs_d = { sel: True for sel in reqs if not sel in omit }
selection = self.selection.require(**reqs_d)
self.reqs = [ sel for sel in reqs if not sel in omit ]
if cutflow:
#
cutflow_reqs_d = {}
for req in reqs:
cutflow_reqs_d.update({req: True})
cutflow.addRow( req, self.selection.require(**cutflow_reqs_d) )
return selection
def process(self, events):
output = self.accumulator.identity()
# use a very loose preselection to filter the events
presel = ak.num(events.Jet) > 2
ev = events[presel]
dataset = ev.metadata['dataset']
# load the config - probably not needed anymore
cfg = loadConfig()
output['totalEvents']['all'] += len(events)
output['skimmedEvents']['all'] += len(ev)
## Muons
muon = Collections(ev, "Muon", "tightTTH").get()
vetomuon = Collections(ev, "Muon", "vetoTTH").get()
dimuon = choose(muon, 2)
SSmuon = ak.any((dimuon['0'].charge * dimuon['1'].charge) > 0, axis=1)
OSmuon = ak.any((dimuon['0'].charge * dimuon['1'].charge) < 0, axis=1)
leading_muon_idx = ak.singletons(ak.argmax(muon.pt, axis=1))
leading_muon = muon[leading_muon_idx]
## Electrons
electron = Collections(ev, "Electron", "tightTTH").get()
vetoelectron = Collections(ev, "Electron", "vetoTTH").get()
dielectron = choose(electron, 2)
SSelectron = ak.any(
(dielectron['0'].charge * dielectron['1'].charge) > 0, axis=1)
OSelectron = ak.any(
(dielectron['0'].charge * dielectron['1'].charge) < 0, axis=1)
leading_electron_idx = ak.singletons(ak.argmax(electron.pt, axis=1))
leading_electron = electron[leading_electron_idx]
## Merge electrons and muons - this should work better now in ak1
lepton = ak.concatenate([muon, electron], axis=1)
dilepton = cross(muon, electron)
SSlepton = ak.any((dilepton['0'].charge * dilepton['1'].charge) > 0,
axis=1)
OSlepton = ak.any((dilepton['0'].charge * dilepton['1'].charge) < 0,
axis=1)
leading_lepton_idx = ak.singletons(ak.argmax(lepton.pt, axis=1))
leading_lepton = lepton[leading_lepton_idx]
trailing_lepton_idx = ak.singletons(ak.argmin(lepton.pt, axis=1))
trailing_lepton = lepton[trailing_lepton_idx]
## Jets
jet = getJets(ev, minPt=25, maxEta=4.7, pt_var='pt_nom')
jet = jet[ak.argsort(
jet.pt_nom, ascending=False
)] # need to sort wrt smeared and recorrected jet pt
jet = jet[~match(jet, muon,
deltaRCut=0.4)] # remove jets that overlap with muons
jet = jet[~match(
jet, electron,
deltaRCut=0.4)] # remove jets that overlap with electrons
central = jet[(abs(jet.eta) < 2.4)]
btag = getBTagsDeepFlavB(
jet, year=self.year) # should study working point for DeepJet
light = getBTagsDeepFlavB(jet, year=self.year, invert=True)
fwd = getFwdJet(light)
fwd_noPU = getFwdJet(light, puId=False)
## forward jets
high_p_fwd = fwd[ak.singletons(ak.argmax(
fwd.p, axis=1))] # highest momentum spectator
high_pt_fwd = fwd[ak.singletons(ak.argmax(
fwd.pt_nom, axis=1))] # highest transverse momentum spectator
high_eta_fwd = fwd[ak.singletons(ak.argmax(abs(
fwd.eta), axis=1))] # most forward spectator
## Get the two leading b-jets in terms of btag score
high_score_btag = central[ak.argsort(central.btagDeepFlavB)][:, :2]
jf = cross(high_p_fwd, jet)
mjf = (jf['0'] + jf['1']).mass
deltaEta = abs(high_p_fwd.eta -
jf[ak.singletons(ak.argmax(mjf, axis=1))]['1'].eta)
deltaEtaMax = ak.max(deltaEta, axis=1)
mjf_max = ak.max(mjf, axis=1)
jj = choose(jet, 2)
mjj_max = ak.max((jj['0'] + jj['1']).mass, axis=1)
## MET -> can switch to puppi MET
met_pt = ev.MET.pt
met_phi = ev.MET.phi
## other variables
ht = ak.sum(jet.pt, axis=1)
st = met_pt + ht + ak.sum(muon.pt, axis=1) + ak.sum(electron.pt,
axis=1)
ht_central = ak.sum(central.pt, axis=1)
## event selectors
filters = getFilters(ev, year=self.year, dataset=dataset)
triggers = getTriggers(ev, year=self.year, dataset=dataset)
dilep = ((ak.num(electron) == 1) & (ak.num(muon) == 1))
lep0pt = ((ak.num(electron[(electron.pt > 25)]) +
ak.num(muon[(muon.pt > 25)])) > 0)
lep1pt = ((ak.num(electron[(electron.pt > 20)]) +
ak.num(muon[(muon.pt > 20)])) > 1)
lepveto = ((ak.num(vetoelectron) + ak.num(vetomuon)) == 2)
# define the weight
weight = Weights(len(ev))
if not dataset == 'MuonEG':
# lumi weight
weight.add("weight", ev.weight * cfg['lumi'][self.year])
# PU weight - not in the babies...
weight.add("PU",
ev.puWeight,
weightUp=ev.puWeightUp,
weightDown=ev.puWeightDown,
shift=False)
# b-tag SFs
weight.add("btag", self.btagSF.Method1a(btag, light))
# lepton SFs
weight.add("lepton", self.leptonSF.get(electron, muon))
selection = PackedSelection()
selection.add('lepveto', lepveto)
selection.add('dilep', dilep)
selection.add('trigger', (triggers))
selection.add('filter', (filters))
selection.add('p_T(lep0)>25', lep0pt)
selection.add('p_T(lep1)>20', lep1pt)
selection.add('OS', OSlepton)
selection.add('N_btag=2', (ak.num(btag) == 2))
selection.add('N_jet>2', (ak.num(jet) >= 3))
selection.add('N_central>1', (ak.num(central) >= 2))
selection.add('N_fwd>0', (ak.num(fwd) >= 1))
selection.add('MET>30', (ev.MET.pt > 30))
os_reqs = [
'lepveto', 'dilep', 'trigger', 'filter', 'p_T(lep0)>25',
'p_T(lep1)>20', 'OS'
]
bl_reqs = os_reqs + [
'N_btag=2', 'N_jet>2', 'N_central>1', 'N_fwd>0', 'MET>30'
]
os_reqs_d = {sel: True for sel in os_reqs}
os_selection = selection.require(**os_reqs_d)
bl_reqs_d = {sel: True for sel in bl_reqs}
BL = selection.require(**bl_reqs_d)
cutflow = Cutflow(output, ev, weight=weight)
cutflow_reqs_d = {}
for req in bl_reqs:
cutflow_reqs_d.update({req: True})
cutflow.addRow(req, selection.require(**cutflow_reqs_d))
# first, make a few super inclusive plots
output['PV_npvs'].fill(dataset=dataset,
multiplicity=ev.PV[os_selection].npvs,
weight=weight.weight()[os_selection])
output['PV_npvsGood'].fill(dataset=dataset,
multiplicity=ev.PV[os_selection].npvsGood,
weight=weight.weight()[os_selection])
output['N_jet'].fill(dataset=dataset,
multiplicity=ak.num(jet)[os_selection],
weight=weight.weight()[os_selection])
output['N_b'].fill(dataset=dataset,
multiplicity=ak.num(btag)[os_selection],
weight=weight.weight()[os_selection])
output['N_central'].fill(dataset=dataset,
multiplicity=ak.num(central)[os_selection],
weight=weight.weight()[os_selection])
output['N_ele'].fill(dataset=dataset,
multiplicity=ak.num(electron)[os_selection],
weight=weight.weight()[os_selection])
output['N_mu'].fill(dataset=dataset,
multiplicity=ak.num(electron)[os_selection],
weight=weight.weight()[os_selection])
output['N_fwd'].fill(dataset=dataset,
multiplicity=ak.num(fwd)[os_selection],
weight=weight.weight()[os_selection])
output['MET'].fill(dataset=dataset,
pt=ev.MET[os_selection].pt,
phi=ev.MET[os_selection].phi,
weight=weight.weight()[os_selection])
output['electron'].fill(dataset=dataset,
pt=ak.to_numpy(ak.flatten(electron[BL].pt)),
eta=ak.to_numpy(ak.flatten(electron[BL].eta)),
phi=ak.to_numpy(ak.flatten(electron[BL].phi)),
weight=weight.weight()[BL])
output['muon'].fill(dataset=dataset,
pt=ak.to_numpy(ak.flatten(muon[BL].pt)),
eta=ak.to_numpy(ak.flatten(muon[BL].eta)),
phi=ak.to_numpy(ak.flatten(muon[BL].phi)),
weight=weight.weight()[BL])
output['lead_lep'].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(leading_lepton[BL].pt)),
eta=ak.to_numpy(ak.flatten(leading_lepton[BL].eta)),
phi=ak.to_numpy(ak.flatten(leading_lepton[BL].phi)),
weight=weight.weight()[BL])
output['trail_lep'].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(trailing_lepton[BL].pt)),
eta=ak.to_numpy(ak.flatten(trailing_lepton[BL].eta)),
phi=ak.to_numpy(ak.flatten(trailing_lepton[BL].phi)),
weight=weight.weight()[BL])
output['fwd_jet'].fill(dataset=dataset,
pt=ak.flatten(high_p_fwd[BL].pt_nom),
eta=ak.flatten(high_p_fwd[BL].eta),
phi=ak.flatten(high_p_fwd[BL].phi),
weight=weight.weight()[BL])
output['b1'].fill(dataset=dataset,
pt=ak.flatten(high_score_btag[:, 0:1][BL].pt_nom),
eta=ak.flatten(high_score_btag[:, 0:1][BL].eta),
phi=ak.flatten(high_score_btag[:, 0:1][BL].phi),
weight=weight.weight()[BL])
output['b2'].fill(dataset=dataset,
pt=ak.flatten(high_score_btag[:, 1:2][BL].pt_nom),
eta=ak.flatten(high_score_btag[:, 1:2][BL].eta),
phi=ak.flatten(high_score_btag[:, 1:2][BL].phi),
weight=weight.weight()[BL])
output['j1'].fill(dataset=dataset,
pt=ak.flatten(jet.pt_nom[:, 0:1][BL]),
eta=ak.flatten(jet.eta[:, 0:1][BL]),
phi=ak.flatten(jet.phi[:, 0:1][BL]),
weight=weight.weight()[BL])
output['j2'].fill(dataset=dataset,
pt=ak.flatten(jet[:, 1:2][BL].pt_nom),
eta=ak.flatten(jet[:, 1:2][BL].eta),
phi=ak.flatten(jet[:, 1:2][BL].phi),
weight=weight.weight()[BL])
output['j3'].fill(dataset=dataset,
pt=ak.flatten(jet[:, 2:3][BL].pt_nom),
eta=ak.flatten(jet[:, 2:3][BL].eta),
phi=ak.flatten(jet[:, 2:3][BL].phi),
weight=weight.weight()[BL])
# Now, take care of systematic unceratinties
if not dataset == 'MuonEG':
alljets = getJets(ev, minPt=0, maxEta=4.7)
alljets = alljets[(alljets.jetId > 1)]
for var in self.variations:
# get the collections that change with the variations
jet = getPtEtaPhi(alljets, pt_var=var)
jet = jet[(jet.pt > 25)]
jet = jet[~match(
jet, muon,
deltaRCut=0.4)] # remove jets that overlap with muons
jet = jet[~match(
jet, electron,
deltaRCut=0.4)] # remove jets that overlap with electrons
central = jet[(abs(jet.eta) < 2.4)]
btag = getBTagsDeepFlavB(
jet,
year=self.year) # should study working point for DeepJet
light = getBTagsDeepFlavB(jet, year=self.year, invert=True)
fwd = getFwdJet(light)
fwd_noPU = getFwdJet(light, puId=False)
## forward jets
high_p_fwd = fwd[ak.singletons(ak.argmax(
fwd.p, axis=1))] # highest momentum spectator
high_pt_fwd = fwd[ak.singletons(ak.argmax(
fwd.pt, axis=1))] # highest transverse momentum spectator
high_eta_fwd = fwd[ak.singletons(
ak.argmax(abs(fwd.eta), axis=1))] # most forward spectator
## Get the two leading b-jets in terms of btag score
high_score_btag = central[ak.argsort(
central.btagDeepFlavB)][:, :2]
# get the modified selection -> more difficult
selection.add('N_jet>2_' + var,
(ak.num(jet.pt) >= 3)) # stupid bug here...
selection.add('N_btag=2_' + var, (ak.num(btag) == 2))
selection.add('N_central>1_' + var, (ak.num(central) >= 2))
selection.add('N_fwd>0_' + var, (ak.num(fwd) >= 1))
selection.add('MET>30_' + var, (getattr(ev.MET, var) > 30))
## Don't change the selection for now...
bl_reqs = os_reqs + [
'N_jet>2_' + var, 'MET>30_' + var, 'N_btag=2_' + var,
'N_central>1_' + var, 'N_fwd>0_' + var
]
bl_reqs_d = {sel: True for sel in bl_reqs}
BL = selection.require(**bl_reqs_d)
# the OS selection remains unchanged
output['N_jet_' + var].fill(
dataset=dataset,
multiplicity=ak.num(jet)[os_selection],
weight=weight.weight()[os_selection])
output['N_fwd_' + var].fill(
dataset=dataset,
multiplicity=ak.num(fwd)[os_selection],
weight=weight.weight()[os_selection])
output['N_b_' + var].fill(
dataset=dataset,
multiplicity=ak.num(btag)[os_selection],
weight=weight.weight()[os_selection])
output['N_central_' + var].fill(
dataset=dataset,
multiplicity=ak.num(central)[os_selection],
weight=weight.weight()[os_selection])
# We don't need to redo all plots with variations. E.g., just add uncertainties to the jet plots.
output['j1_' + var].fill(dataset=dataset,
pt=ak.flatten(jet.pt[:, 0:1][BL]),
eta=ak.flatten(jet.eta[:, 0:1][BL]),
phi=ak.flatten(jet.phi[:, 0:1][BL]),
weight=weight.weight()[BL])
output['b1_' + var].fill(
dataset=dataset,
pt=ak.flatten(high_score_btag[:, 0:1].pt[:, 0:1][BL]),
eta=ak.flatten(high_score_btag[:, 0:1].eta[:, 0:1][BL]),
phi=ak.flatten(high_score_btag[:, 0:1].phi[:, 0:1][BL]),
weight=weight.weight()[BL])
output['fwd_jet_' + var].fill(
dataset=dataset,
pt=ak.flatten(high_p_fwd[BL].pt),
eta=ak.flatten(high_p_fwd[BL].eta),
phi=ak.flatten(high_p_fwd[BL].phi),
weight=weight.weight()[BL])
output['MET_' + var].fill(dataset=dataset,
pt=getattr(ev.MET,
var)[os_selection],
phi=ev.MET[os_selection].phi,
weight=weight.weight()[os_selection])
return output
def process(self, events):
output = self.accumulator.identity()
# we can use a very loose preselection to filter the events. nothing is done with this presel, though
presel = ak.num(events.Jet)>=2
ev = events[presel]
dataset = ev.metadata['dataset']
# load the config - probably not needed anymore
cfg = loadConfig()
output['totalEvents']['all'] += len(events)
output['skimmedEvents']['all'] += len(ev)
## Electrons
electron = Collections(ev, "Electron", "tightFCNC", 0, self.year).get()
electron = electron[(electron.pt > 15) & (np.abs(electron.eta) < 2.4)]
electron = electron[(electron.genPartIdx >= 0)]
electron = electron[(np.abs(electron.matched_gen.pdgId)==11)] #from here on all leptons are gen-matched
electron = electron[( (electron.genPartFlav==1) | (electron.genPartFlav==15) )] #and now they are all prompt
leading_electron_idx = ak.singletons(ak.argmax(electron.pt, axis=1))
leading_electron = electron[leading_electron_idx]
trailing_electron_idx = ak.singletons(ak.argmin(electron.pt, axis=1))
trailing_electron = electron[trailing_electron_idx]
leading_parent = find_first_parent(leading_electron.matched_gen)
trailing_parent = find_first_parent(trailing_electron.matched_gen)
is_flipped = ( ( (electron.matched_gen.pdgId*(-1) == electron.pdgId) | (find_first_parent(electron.matched_gen)*(-1) == electron.pdgId) ) & (np.abs(electron.pdgId) == 11) )
flipped_electron = electron[is_flipped]
flipped_electron = flipped_electron[(ak.fill_none(flipped_electron.pt, 0)>0)]
flipped_electron = flipped_electron[~(ak.is_none(flipped_electron))]
n_flips = ak.num(flipped_electron)
##Muons
muon = Collections(ev, "Muon", "tightFCNC").get()
muon = muon[(muon.pt > 15) & (np.abs(muon.eta) < 2.4)]
muon = muon[(muon.genPartIdx >= 0)]
muon = muon[(np.abs(muon.matched_gen.pdgId)==13)] #from here, all muons are gen-matched
muon = muon[( (muon.genPartFlav==1) | (muon.genPartFlav==15) )] #and now they are all prompt
##Leptons
lepton = ak.concatenate([muon, electron], axis=1)
SSlepton = (ak.sum(lepton.charge, axis=1) != 0) & (ak.num(lepton)==2)
OSlepton = (ak.sum(lepton.charge, axis=1) == 0) & (ak.num(lepton)==2)
emulepton = (ak.num(electron) == 1) & (ak.num(muon) == 1)
no_mumu = (ak.num(muon) <= 1)
leading_lepton_idx = ak.singletons(ak.argmax(lepton.pt, axis=1))
leading_lepton = lepton[leading_lepton_idx]
trailing_lepton_idx = ak.singletons(ak.argmin(lepton.pt, axis=1))
trailing_lepton = lepton[trailing_lepton_idx]
#jets
jet = getJets(ev, minPt=40, maxEta=2.4, pt_var='pt')
jet = jet[ak.argsort(jet.pt, ascending=False)] # need to sort wrt smeared and recorrected jet pt
jet = jet[~match(jet, muon, deltaRCut=0.4)] # remove jets that overlap with muons
jet = jet[~match(jet, electron, deltaRCut=0.4)]
## MET -> can switch to puppi MET
met_pt = ev.MET.pt
met_phi = ev.MET.phi
# setting up the various weights
weight = Weights( len(ev) )
weight2 = Weights( len(ev))
if not dataset=='MuonEG':
# generator weight
weight.add("weight", ev.genWeight)
weight2.add("weight", ev.genWeight)
weight2.add("charge flip", self.charge_flip_ratio.flip_weight(electron))
#selections
filters = getFilters(ev, year=self.year, dataset=dataset)
ss = (SSlepton)
os = (OSlepton)
jet_all = (ak.num(jet) >= 2)
diele = (ak.num(electron) == 2)
emu = (emulepton)
flips = (n_flips == 1)
no_flips = (n_flips == 0)
nmm = no_mumu
selection = PackedSelection()
selection.add('filter', (filters) )
selection.add('ss', ss )
selection.add('os', os )
selection.add('jet', jet_all )
selection.add('ee', diele)
selection.add('emu', emu)
selection.add('flip', flips)
selection.add('nflip', no_flips)
selection.add('no_mumu', nmm)
bl_reqs = ['filter'] + ['jet']
bl_reqs_d = { sel: True for sel in bl_reqs }
baseline = selection.require(**bl_reqs_d)
f_reqs = bl_reqs + ['flip'] + ['ss'] + ['ee']
f_reqs_d = {sel: True for sel in f_reqs}
flip_sel = selection.require(**f_reqs_d)
f2_reqs = bl_reqs + ['flip'] + ['ss'] + ['emu']
f2_reqs_d = {sel: True for sel in f2_reqs}
flip_sel2 = selection.require(**f2_reqs_d)
f3_reqs = bl_reqs + ['flip'] + ['ss'] + ['no_mumu']
f3_reqs_d = {sel: True for sel in f3_reqs}
flip_sel3 = selection.require(**f3_reqs_d)
nf_reqs = bl_reqs + ['nflip'] + ['os'] + ['ee']
nf_reqs_d = {sel: True for sel in nf_reqs}
n_flip_sel = selection.require(**nf_reqs_d)
nf2_reqs = bl_reqs + ['nflip'] + ['os'] + ['emu']
nf2_reqs_d = {sel: True for sel in nf2_reqs}
n_flip_sel2 = selection.require(**nf2_reqs_d)
nf3_reqs = bl_reqs + ['nflip'] + ['os'] + ['no_mumu']
nf3_reqs_d = {sel: True for sel in nf3_reqs}
n_flip_sel3 = selection.require(**nf3_reqs_d)
s_reqs = bl_reqs + ['ss'] + ['no_mumu']
s_reqs_d = { sel: True for sel in s_reqs }
ss_sel = selection.require(**s_reqs_d)
o_reqs = bl_reqs + ['os'] + ['no_mumu']
o_reqs_d = {sel: True for sel in o_reqs }
os_sel = selection.require(**o_reqs_d)
ees_reqs = bl_reqs + ['ss'] + ['ee']
ees_reqs_d = { sel: True for sel in ees_reqs }
eess_sel = selection.require(**ees_reqs_d)
eeo_reqs = bl_reqs + ['os'] + ['ee']
eeo_reqs_d = {sel: True for sel in eeo_reqs }
eeos_sel = selection.require(**eeo_reqs_d)
ems_reqs = bl_reqs + ['ss'] + ['emu']
ems_reqs_d = { sel: True for sel in ems_reqs }
emss_sel = selection.require(**ems_reqs_d)
emo_reqs = bl_reqs + ['os'] + ['emu']
emo_reqs_d = {sel: True for sel in emo_reqs }
emos_sel = selection.require(**emo_reqs_d)
#outputs
output['N_jet'].fill(dataset=dataset, multiplicity=ak.num(jet)[baseline], weight=weight.weight()[baseline])
output['N_ele'].fill(dataset=dataset, multiplicity=ak.num(lepton)[ss_sel], weight=weight.weight()[ss_sel])
output['N_ele2'].fill(dataset=dataset, multiplicity=ak.num(lepton)[os_sel], weight=weight2.weight()[os_sel])
output['electron_flips'].fill(dataset=dataset, multiplicity = n_flips[flip_sel], weight=weight.weight()[flip_sel])
output['electron_flips2'].fill(dataset=dataset, multiplicity = n_flips[n_flip_sel], weight=weight2.weight()[n_flip_sel])
output['electron_flips3'].fill(dataset=dataset, multiplicity = n_flips[flip_sel2], weight=weight.weight()[flip_sel2])
output['electron_flips4'].fill(dataset=dataset, multiplicity = n_flips[n_flip_sel2], weight=weight2.weight()[n_flip_sel2])
output["electron"].fill(
dataset = dataset,
pt = ak.to_numpy(ak.flatten(leading_electron[flip_sel3].pt)),
eta = np.abs(ak.to_numpy(ak.flatten(leading_electron[flip_sel3].eta))),
weight = weight.weight()[flip_sel3]
)
output["electron2"].fill(
dataset = dataset,
pt = ak.to_numpy(ak.flatten(leading_electron[n_flip_sel3].pt)),
eta = np.abs(ak.to_numpy(ak.flatten(leading_electron[n_flip_sel3].eta))),
weight = weight2.weight()[n_flip_sel3]
)
output["flipped_electron"].fill(
dataset = dataset,
pt = ak.to_numpy(ak.flatten(leading_electron[flip_sel].pt)),
eta = np.abs(ak.to_numpy(ak.flatten(leading_electron[flip_sel].eta))),
weight = weight.weight()[flip_sel]
)
output["flipped_electron2"].fill(
dataset = dataset,
pt = ak.to_numpy(ak.flatten(leading_electron[n_flip_sel].pt)),
eta = np.abs(ak.to_numpy(ak.flatten(leading_electron[n_flip_sel].eta))),
weight = weight2.weight()[n_flip_sel]
)
output["flipped_electron3"].fill(
dataset = dataset,
pt = ak.to_numpy(ak.flatten(leading_electron[flip_sel2].pt)),
eta = np.abs(ak.to_numpy(ak.flatten(leading_electron[flip_sel2].eta))),
weight = weight.weight()[flip_sel2]
)
output["flipped_electron4"].fill(
dataset = dataset,
pt = ak.to_numpy(ak.flatten(leading_electron[n_flip_sel2].pt)),
eta = np.abs(ak.to_numpy(ak.flatten(leading_electron[n_flip_sel2].eta))),
weight = weight2.weight()[n_flip_sel2]
)
#output["lepton_parent"].fill(
# dataset = dataset,
# pdgID = np.abs(ak.to_numpy(ak.flatten(leading_parent[ss_sel]))),
# weight = weight.weight()[ss_sel]
#)
#
#output["lepton_parent2"].fill(
# dataset = dataset,
# pdgID = np.abs(ak.to_numpy(ak.flatten(trailing_parent[ss_sel]))),
# weight = weight.weight()[ss_sel]
#)
return output
def process(self, events):
events = events[
ak.num(events.Jet) >
0] #corrects for rare case where there isn't a single jet in event
output = self.accumulator.identity()
# we can use a very loose preselection to filter the events. nothing is done with this presel, though
presel = ak.num(events.Jet) >= 0
ev = events[presel]
dataset = ev.metadata['dataset']
# load the config - probably not needed anymore
# cfg = loadConfig()
output['totalEvents']['all'] += len(events)
output['skimmedEvents']['all'] += len(ev)
### For FCNC, we want electron -> tightTTH
electron = Collections(ev, "Electron", "tightFCNC").get()
fakeableelectron = Collections(ev, "Electron", "fakeableFCNC").get()
muon = Collections(ev, "Muon", "tightFCNC").get()
fakeablemuon = Collections(ev, "Muon", "fakeableFCNC").get()
##Jets
Jets = events.Jet
## MET -> can switch to puppi MET
met_pt = ev.MET.pt
met_phi = ev.MET.phi
lepton = fakeablemuon #ak.concatenate([fakeablemuon, fakeableelectron], axis=1)
mt_lep_met = mt(lepton.pt, lepton.phi, ev.MET.pt, ev.MET.phi)
min_mt_lep_met = ak.min(mt_lep_met, axis=1)
selection = PackedSelection()
selection.add('MET<20', (ev.MET.pt < 20))
selection.add('mt<20', min_mt_lep_met < 20)
#selection.add('MET<19', (ev.MET.pt<19) )
selection_reqs = ['MET<20', 'mt<20'] #, 'MET<19']
fcnc_reqs_d = {sel: True for sel in selection_reqs}
fcnc_selection = selection.require(**fcnc_reqs_d)
# define the weight
weight = Weights(len(ev))
if not dataset == 'MuonEG':
# generator weight
weight.add("weight", ev.genWeight)
jets = getJets(
ev, maxEta=2.4, minPt=25, pt_var='pt'
) #& (ak.num(jets[~match(jets, fakeablemuon, deltaRCut=1.0)])>=1)
single_muon_sel = (ak.num(muon) == 1) & (ak.num(fakeablemuon) == 1) | (
ak.num(muon) == 0) & (ak.num(fakeablemuon) == 1)
single_electron_sel = (ak.num(electron) == 1) & (
ak.num(fakeableelectron)
== 1) | (ak.num(electron) == 0) & (ak.num(fakeableelectron) == 1)
fcnc_muon_sel = (ak.num(
jets[~match(jets, fakeablemuon, deltaRCut=1.0)]) >=
1) & fcnc_selection & single_muon_sel
fcnc_electron_sel = (ak.num(
jets[~match(jets, fakeableelectron, deltaRCut=1.0)]) >=
1) & fcnc_selection & single_electron_sel
tight_muon_sel = (ak.num(muon) == 1) & fcnc_muon_sel
loose_muon_sel = (ak.num(fakeablemuon) == 1) & fcnc_muon_sel
tight_electron_sel = (ak.num(electron) == 1) & fcnc_electron_sel
loose_electron_sel = (ak.num(fakeableelectron)
== 1) & fcnc_electron_sel
output['single_mu_fakeable'].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(fakeablemuon[loose_muon_sel].conePt)),
eta=np.abs(
ak.to_numpy(ak.flatten(fakeablemuon[loose_muon_sel].eta))))
output['single_mu'].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(muon[tight_muon_sel].conePt)),
eta=np.abs(ak.to_numpy(ak.flatten(muon[tight_muon_sel].eta))))
output['single_e_fakeable'].fill(
dataset=dataset,
pt=ak.to_numpy(
ak.flatten(fakeableelectron[loose_electron_sel].conePt)),
eta=np.abs(
ak.to_numpy(
ak.flatten(fakeableelectron[loose_electron_sel].eta))))
output['single_e'].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(electron[tight_electron_sel].conePt)),
eta=np.abs(
ak.to_numpy(ak.flatten(electron[tight_electron_sel].eta))))
if self.debug:
#create pandas dataframe for debugging
passed_events = ev[tight_muon_sel]
passed_muons = muon[tight_muon_sel]
event_p = ak.to_pandas(passed_events[["event"]])
event_p["MET_PT"] = passed_events["MET"]["pt"]
event_p["mt"] = min_mt_lep_met[tight_muon_sel]
event_p["num_tight_mu"] = ak.to_numpy(ak.num(muon)[tight_muon_sel])
event_p["num_loose_mu"] = ak.num(fakeablemuon)[tight_muon_sel]
muon_p = ak.to_pandas(
ak.flatten(passed_muons)[[
"pt", "conePt", "eta", "dz", "dxy", "ptErrRel",
"miniPFRelIso_all", "jetRelIsoV2", "jetRelIso",
"jetPtRelv2"
]])
#convert to numpy array for the output
events_array = pd.concat([muon_p, event_p], axis=1)
events_to_add = [6886009]
for e in events_to_add:
tmp_event = ev[ev.event == e]
added_event = ak.to_pandas(tmp_event[["event"]])
added_event["MET_PT"] = tmp_event["MET"]["pt"]
added_event["mt"] = min_mt_lep_met[ev.event == e]
added_event["num_tight_mu"] = ak.to_numpy(
ak.num(muon)[ev.event == e])
added_event["num_loose_mu"] = ak.to_numpy(
ak.num(fakeablemuon)[ev.event == e])
add_muon = ak.to_pandas(
ak.flatten(muon[ev.event == e])[[
"pt", "conePt", "eta", "dz", "dxy", "ptErrRel",
"miniPFRelIso_all", "jetRelIsoV2", "jetRelIso",
"jetPtRelv2"
]])
add_concat = pd.concat([add_muon, added_event], axis=1)
events_array = pd.concat([events_array, add_concat], axis=0)
output['muons_df'] += processor.column_accumulator(
events_array.to_numpy())
return output
def process(self, events):
output = self._accumulator.identity()
dataset_name = events.metadata['dataset']
output["total_events"][dataset_name] += events.__len__()
# HLT selection
HLT_mask = []
if year == "2016":
if "SingleMuon" in dataset_name:
if "2016B2" in dataset_name:
HLT_mask = events.HLT.IsoMu24 | events.HLT.IsoTkMu24 | events.HLT.Mu50
else:
HLT_mask = events.HLT.IsoMu24 | events.HLT.IsoTkMu24 | events.HLT.Mu50 | events.HLT.TkMu50
else: #https://twiki.cern.ch/twiki/bin/view/CMS/HLTPathsRunIIList
if "2016B2" in dataset_name:
HLT_mask = events.HLT.PFHT800 | events.HLT.PFHT900 | events.HLT.PFJet500 | events.HLT.CaloJet500_NoJetID
elif "2016H" in dataset_name:
HLT_mask = events.HLT.PFHT900 | events.HLT.AK8PFJet450 | events.HLT.AK8PFJet500 | events.HLT.PFJet500 | events.HLT.CaloJet500_NoJetID
else:
HLT_mask = events.HLT.PFHT800 | events.HLT.PFHT900 | events.HLT.AK8PFJet450 | events.HLT.AK8PFJet500 | events.HLT.PFJet500 | events.HLT.CaloJet500_NoJetID
if year == "2017":
if "SingleMuon" in dataset_name:
if "2017B" in dataset_name:
HLT_mask = events.HLT.IsoMu27 | events.HLT.Mu50
else:
HLT_mask = events.HLT.IsoMu27 | events.HLT.Mu50 | events.HLT.OldMu100 | events.HLT.TkMu100
else:
HLT_mask = events.HLT.PFHT1050 | events.HLT.AK8PFJet500 | events.HLT.AK8PFJet550 | events.HLT.CaloJet500_NoJetID | events.HLT.CaloJet550_NoJetID | events.HLT.PFJet500
if year == "2018":
if "SingleMuon" in dataset_name:
HLT_mask = events.HLT.IsoMu24 | events.HLT.Mu50 | events.HLT.OldMu100 | events.HLT.TkMu100
else:
HLT_mask = events.HLT.PFHT1050 | events.HLT.AK8PFJet500 | events.HLT.AK8PFJet550 | events.HLT.CaloJet500_NoJetID | events.HLT.CaloJet550_NoJetID | events.HLT.PFJet500
# Require 3 jets
jet_mask = (events.Jet.pt > 30.) & (abs(events.Jet.eta) <
2.5) & (events.Jet.isTight)
event_mask = (awk.sum(jet_mask, axis=1) >= 3)
event_mask = event_mask & HLT_mask
events_3j = events[event_mask]
# Reduce jet mask to only events with 3 good jets
jet_mask = jet_mask[event_mask]
# Array of the jets to consider for trijet resonance
selected_jets = events_3j.Jet[jet_mask][:, :3]
# Pairs of jets
pairs = [(0, 1), (1, 2), (2, 0)]
jet_i, jet_j = zip(*pairs) # Returns [0, 1, 2] , [1, 2, 0]
m_ij = (selected_jets[:, jet_i] + selected_jets[:, jet_j]).mass
dR_ij = selected_jets[:, jet_i].delta_r(selected_jets[:, jet_j])
dEta_ij = abs(selected_jets[:, jet_i].eta -
selected_jets[:, jet_j].eta)
max_dR = awk.max(dR_ij, axis=1)
max_dEta = awk.max(dEta_ij, axis=1)
min_dR = awk.min(dR_ij, axis=1)
min_dEta = awk.min(dEta_ij, axis=1)
min_pT = awk.min(selected_jets.pt, axis=1)
max_eta = abs(awk.max(selected_jets.eta, axis=1))
jet_k = [2, 0, 1]
dR_i_jk = selected_jets[:, jet_i].delta_r(selected_jets[:, jet_j] +
selected_jets[:, jet_k])
dEta_i_jk = abs(selected_jets[:, jet_i].eta -
(selected_jets[:, jet_j] +
selected_jets[:, jet_k]).eta)
dPhi_i_jk = abs(selected_jets[:, jet_i].phi -
(selected_jets[:, jet_j] +
selected_jets[:, jet_k]).phi)
dPt_i_jk = abs(selected_jets[:, jet_i].pt -
(selected_jets[:, jet_j] + selected_jets[:, jet_k]).pt)
max_dPhi_jjj = awk.max(dPhi_i_jk, axis=1)
m3j = selected_jets.sum().mass
pt_i_overM = selected_jets.pt / m3j
max_pt_overM = awk.max(pt_i_overM, axis=1)
min_pt_overM = awk.min(pt_i_overM, axis=1)
m_01_overM = m_ij[:, 0] / m3j
m_12_overM = m_ij[:, 1] / m3j
m_20_overM = m_ij[:, 2] / m3j
for pt_cut in range(30, 1150, 5):
cut_name = f"min_pT_cut{pt_cut}".format(pt_cut)
selection = PackedSelection()
selection.add("MinJetPt_cut", min_pT > pt_cut)
sel_mask = selection.require(
**{name: True
for name in selection.names})
output[f"N_min_pT_cut{pt_cut}".format(
pt_cut)][dataset_name] += events_3j[sel_mask].__len__()
for eta_cut in np.arange(0, 2.5, 0.05):
cut_name = f"max_eta_cut{eta_cut}".format(eta_cut)
selection = PackedSelection()
selection.add("MaxJetEta_cut", max_eta < eta_cut)
sel_mask = selection.require(
**{name: True
for name in selection.names})
output[f"N_max_eta_cut{eta_cut}".format(
eta_cut)][dataset_name] += events_3j[sel_mask].__len__()
for dEta_max_cut in np.arange(0, 5, 0.1):
cut_name = f"dEta_max_cut{dEta_max_cut}".format(dEta_max_cut)
selection = PackedSelection()
selection.add("MaxJJdEta_cut", max_dEta < dEta_max_cut)
sel_mask = selection.require(
**{name: True
for name in selection.names})
output[f"N_dEta_jj_max_cut{dEta_max_cut}".format(
dEta_max_cut)][dataset_name] += events_3j[sel_mask].__len__()
for dR_min_cut in np.arange(0, 5, 0.1):
cut_name = f"dR_min_cut{dR_min_cut}".format(dR_min_cut)
selection = PackedSelection()
selection.add("MinJJdR_cut", min_dR > dR_min_cut)
sel_mask = selection.require(
**{name: True
for name in selection.names})
output[f"N_dR_jj_min_cut{dR_min_cut}".format(
dR_min_cut)][dataset_name] += events_3j[sel_mask].__len__()
#min cut for the variable dPhi_jjj_max
# for dPhi_jjj_max_min_cut in range(0,6,0.1):
# cut_name = f"dPhi_jjj_max_min_cut{dPhi_jjj_max_min_cut}".format(dPhi_jjj_max_min_cut)
# selections[cut_name] = PackedSelection()
# selections[cut_name].add("j_jj_dPhi_max_cut", min_dR > dPhi_jjj_max_min_cut)
# selection_items[cut_name] = []
# selection_items[cut_name].append("j_jj_dPhi_max_cut")
# sel_mask = HLT_mask & selections[cut_name].require(**{name: True for name in selection_items[cut_name]})
# output[f"N_dPhi_jjj_max_min_cut{dPhi_jjj_max_min_cut}".format(dPhi_jjj_max_min_cut)][dataset_name] += events_3j[sel_mask].__len__()
# for dPhi_jjj_min_max_cut in range(0,6,0.1):
# cut_name = f"dPhi_jjj_max_min_cut{dPhi_jjj_max_min_cut}".format(dPhi_jjj_max_min_cut)
# selections[cut_name] = PackedSelection()
# selections[cut_name].add("j_jj_dPhi_max_cut", min_dR > dPhi_jjj_max_min_cut)
# selection_items[cut_name] = []
# selection_items[cut_name].append("j_jj_dPhi_max_cut")
# sel_mask = HLT_mask & selections[cut_name].require(**{name: True for name in selection_items[cut_name]})
# output[f"N_dPhi_jjj_max_min_cut{dPhi_jjj_max_min_cut}".format(dPhi_jjj_max_min_cut)][dataset_name] += events_3j[sel_mask].__len__()
return output
def process(self, events):
output = self.accumulator.identity()
# we can use a very loose preselection to filter the events. nothing is done with this presel, though
presel = ak.num(events.Jet) > 0
ev = events[presel]
dataset = ev.metadata['dataset']
# load the config - probably not needed anymore
cfg = loadConfig()
#output['totalEvents']['all'] += len(events)
#output['skimmedEvents']['all'] += len(ev)
if self.year == 2018:
triggers = ev.HLT.Ele23_Ele12_CaloIdL_TrackIdL_IsoVL
elif self.year == 2017:
triggers = ev.HLT.Ele23_Ele12_CaloIdL_TrackIdL_IsoVL
elif self.year == 2016:
triggers = ev.HLT.Ele23_Ele12_CaloIdL_TrackIdL_IsoVL_DZ
if self.year == 2018:
lumimask = LumiMask(
'processors/Cert_314472-325175_13TeV_Legacy2018_Collisions18_JSON.txt'
)
## Electrons
electron = Collections(ev, "Electron", "tight").get()
electron = electron[(electron.pt > 25) & (np.abs(electron.eta) < 2.4)]
loose_electron = Collections(ev, "Electron", "veto").get()
loose_electron = loose_electron[(loose_electron.pt > 25)
& (np.abs(loose_electron.eta) < 2.4)]
SSelectron = (ak.sum(electron.charge, axis=1) != 0) & (ak.num(electron)
== 2)
OSelectron = (ak.sum(electron.charge, axis=1) == 0) & (ak.num(electron)
== 2)
dielectron = choose(electron, 2)
dielectron_mass = (dielectron['0'] + dielectron['1']).mass
dielectron_pt = (dielectron['0'] + dielectron['1']).pt
leading_electron_idx = ak.singletons(ak.argmax(electron.pt, axis=1))
leading_electron = electron[(leading_electron_idx)]
leading_electron = leading_electron[(leading_electron.pt > 30)]
trailing_electron_idx = ak.singletons(ak.argmin(electron.pt, axis=1))
trailing_electron = electron[trailing_electron_idx]
##Muons
loose_muon = Collections(ev, "Muon", "veto").get()
loose_muon = loose_muon[(loose_muon.pt > 20)
& (np.abs(loose_muon.eta) < 2.4)]
#jets
jet = getJets(ev, minPt=40, maxEta=2.4, pt_var='pt', UL=False)
jet = jet[ak.argsort(
jet.pt, ascending=False
)] # need to sort wrt smeared and recorrected jet pt
jet = jet[~match(jet, loose_muon,
deltaRCut=0.4)] # remove jets that overlap with muons
jet = jet[~match(
jet, electron,
deltaRCut=0.4)] # remove jets that overlap with electrons
## MET -> can switch to puppi MET
met_pt = ev.MET.pt
met_phi = ev.MET.phi
#selections
filters = getFilters(ev, year=self.year, dataset=dataset)
mask = lumimask(ev.run, ev.luminosityBlock)
ss = (SSelectron)
os = (OSelectron)
mass = (ak.min(np.abs(dielectron_mass - 91.2), axis=1) < 15)
lead_electron = (ak.min(leading_electron.pt, axis=1) > 30)
jet1 = (ak.num(jet) >= 1)
jet2 = (ak.num(jet) >= 2)
num_loose = ((ak.num(loose_electron) == 2) & (ak.num(loose_muon) == 0))
selection = PackedSelection()
selection.add('filter', (filters))
selection.add('mask', (mask))
selection.add('ss', ss)
selection.add('os', os)
selection.add('mass', mass)
selection.add('leading', lead_electron)
selection.add('triggers', triggers)
selection.add('one jet', jet1)
selection.add('two jets', jet2)
selection.add('num_loose', num_loose)
bl_reqs = ['filter'] + ['mass'] + ['mask'] + ['triggers'] + [
'leading'
] + ['num_loose']
#bl_reqs = ['filter'] + ['mass'] + ['triggers'] + ['leading'] + ['num_loose']
bl_reqs_d = {sel: True for sel in bl_reqs}
baseline = selection.require(**bl_reqs_d)
s_reqs = bl_reqs + ['ss']
s_reqs_d = {sel: True for sel in s_reqs}
ss_sel = selection.require(**s_reqs_d)
o_reqs = bl_reqs + ['os']
o_reqs_d = {sel: True for sel in o_reqs}
os_sel = selection.require(**o_reqs_d)
j1s_reqs = s_reqs + ['one jet']
j1s_reqs_d = {sel: True for sel in j1s_reqs}
j1ss_sel = selection.require(**j1s_reqs_d)
j1o_reqs = o_reqs + ['one jet']
j1o_reqs_d = {sel: True for sel in j1o_reqs}
j1os_sel = selection.require(**j1o_reqs_d)
j2s_reqs = s_reqs + ['two jets']
j2s_reqs_d = {sel: True for sel in j2s_reqs}
j2ss_sel = selection.require(**j2s_reqs_d)
j2o_reqs = o_reqs + ['two jets']
j2o_reqs_d = {sel: True for sel in j2o_reqs}
j2os_sel = selection.require(**j2o_reqs_d)
output["N_jet"].fill(
dataset=dataset,
multiplicity=ak.num(jet)[os_sel],
)
return output
def process(self, events):
output = self.accumulator.identity()
# we can use a very loose preselection to filter the events. nothing is done with this presel, though
presel = ak.num(events.Jet) >= 2
ev = events[presel]
dataset = ev.metadata['dataset']
# load the config - probably not needed anymore
cfg = loadConfig()
output['totalEvents']['all'] += len(events)
output['skimmedEvents']['all'] += len(ev)
## Electrons
electron = Collections(ev, "Electron", "tight").get()
electron = electron[(electron.pt > 20) & (abs(electron.eta) < 2.4)]
electron = electron[((electron.genPartIdx >= 0) &
(np.abs(electron.matched_gen.pdgId) == 11)
)] #from here on all leptons are gen-matched
##Muons
muon = Collections(ev, "Muon", "tight").get()
muon = muon[(muon.pt > 20) & (abs(muon.eta) < 2.4)]
muon = muon[((muon.genPartIdx >= 0) &
(np.abs(muon.matched_gen.pdgId) == 13))]
##Leptons
lepton = ak.concatenate([muon, electron], axis=1)
SSlepton = (ak.sum(lepton.charge, axis=1) != 0) & (ak.num(lepton) == 2)
OSlepton = (ak.sum(lepton.charge, axis=1) == 0) & (ak.num(lepton) == 2)
leading_lepton_idx = ak.singletons(ak.argmax(lepton.pt, axis=1))
leading_lepton = lepton[leading_lepton_idx]
#jets
jet = getJets(ev, minPt=40, maxEta=2.4, pt_var='pt')
jet = jet[ak.argsort(
jet.pt, ascending=False
)] # need to sort wrt smeared and recorrected jet pt
jet = jet[~match(jet, muon,
deltaRCut=0.4)] # remove jets that overlap with muons
jet = jet[~match(jet, electron, deltaRCut=0.4)]
## MET -> can switch to puppi MET
met_pt = ev.MET.pt
met_phi = ev.MET.phi
# setting up the various weights
weight = Weights(len(ev))
weight2 = Weights(len(ev))
if not dataset == 'MuonEG':
# generator weight
weight.add("weight", ev.genWeight)
weight2.add("weight", ev.genWeight)
weight2.add("charge flip",
self.charge_flip_ratio.flip_weight(electron))
#selections
filters = getFilters(ev, year=self.year, dataset=dataset)
ss = (SSlepton)
os = (OSlepton)
jet_all = (ak.num(jet) >= 2)
selection = PackedSelection()
selection.add('filter', (filters))
selection.add('ss', ss)
selection.add('os', os)
selection.add('jet', jet_all)
bl_reqs = ['filter', 'jet']
bl_reqs_d = {sel: True for sel in bl_reqs}
baseline = selection.require(**bl_reqs_d)
s_reqs = bl_reqs + ['ss']
s_reqs_d = {sel: True for sel in s_reqs}
ss_sel = selection.require(**s_reqs_d)
o_reqs = bl_reqs + ['os']
o_reqs_d = {sel: True for sel in o_reqs}
os_sel = selection.require(**o_reqs_d)
#outputs
output['N_jet'].fill(dataset=dataset,
multiplicity=ak.num(jet)[baseline],
weight=weight.weight()[baseline])
output['N_ele'].fill(dataset=dataset,
multiplicity=ak.num(lepton)[ss_sel],
weight=weight.weight()[ss_sel])
output['N_ele2'].fill(dataset=dataset,
multiplicity=ak.num(lepton)[os_sel],
weight=weight2.weight()[os_sel])
output["electron"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(leading_lepton[ss_sel].pt)),
eta=abs(ak.to_numpy(ak.flatten(leading_lepton[ss_sel].eta))),
phi=ak.to_numpy(ak.flatten(leading_lepton[ss_sel].phi)),
weight=weight.weight()[ss_sel])
output["electron2"].fill(
dataset=dataset,
pt=ak.to_numpy(ak.flatten(leading_lepton[os_sel].pt)),
eta=abs(ak.to_numpy(ak.flatten(leading_lepton[os_sel].eta))),
phi=ak.to_numpy(ak.flatten(leading_lepton[os_sel].phi)),
weight=weight2.weight()[os_sel])
return output
def process_shift(self, events, shift_name):
dataset = events.metadata['dataset']
isRealData = not hasattr(events, "genWeight")
selection = PackedSelection()
weights = Weights(len(events), storeIndividual=True)
output = self.make_output()
if shift_name is None and not isRealData:
output['sumw'] = ak.sum(events.genWeight)
if isRealData or self._newTrigger:
trigger = np.zeros(len(events), dtype='bool')
for t in self._triggers[self._year]:
if t in events.HLT.fields:
trigger = trigger | events.HLT[t]
selection.add('trigger', trigger)
del trigger
else:
selection.add('trigger', np.ones(len(events), dtype='bool'))
if isRealData:
selection.add(
'lumimask', lumiMasks[self._year](events.run,
events.luminosityBlock))
else:
selection.add('lumimask', np.ones(len(events), dtype='bool'))
if isRealData and self._skipRunB and self._year == '2017':
selection.add('dropB', events.run > 299329)
else:
selection.add('dropB', np.ones(len(events), dtype='bool'))
if isRealData:
trigger = np.zeros(len(events), dtype='bool')
for t in self._muontriggers[self._year]:
if t in events.HLT.fields:
trigger |= np.array(events.HLT[t])
selection.add('muontrigger', trigger)
del trigger
else:
selection.add('muontrigger', np.ones(len(events), dtype='bool'))
metfilter = np.ones(len(events), dtype='bool')
for flag in self._met_filters[
self._year]['data' if isRealData else 'mc']:
metfilter &= np.array(events.Flag[flag])
selection.add('metfilter', metfilter)
del metfilter
fatjets = events.FatJet
fatjets['msdcorr'] = corrected_msoftdrop(fatjets)
fatjets['qcdrho'] = 2 * np.log(fatjets.msdcorr / fatjets.pt)
fatjets['n2ddt'] = fatjets.n2b1 - n2ddt_shift(fatjets, year=self._year)
fatjets['msdcorr_full'] = fatjets['msdcorr'] * self._msdSF[self._year]
candidatejet = fatjets[
# https://github.com/DAZSLE/BaconAnalyzer/blob/master/Analyzer/src/VJetLoader.cc#L269
(fatjets.pt > 200)
& (abs(fatjets.eta) < 2.5)
& fatjets.isTight # this is loose in sampleContainer
]
candidatejet = candidatejet[:, :
2] # Only consider first two to match generators
if self._jet_arbitration == 'pt':
candidatejet = ak.firsts(candidatejet)
elif self._jet_arbitration == 'mass':
candidatejet = ak.firsts(candidatejet[ak.argmax(
candidatejet.msdcorr, axis=1, keepdims=True)])
elif self._jet_arbitration == 'n2':
candidatejet = ak.firsts(candidatejet[ak.argmin(candidatejet.n2ddt,
axis=1,
keepdims=True)])
elif self._jet_arbitration == 'ddb':
candidatejet = ak.firsts(candidatejet[ak.argmax(
candidatejet.btagDDBvLV2, axis=1, keepdims=True)])
elif self._jet_arbitration == 'ddc':
candidatejet = ak.firsts(candidatejet[ak.argmax(
candidatejet.btagDDCvLV2, axis=1, keepdims=True)])
else:
raise RuntimeError("Unknown candidate jet arbitration")
if self._tagger == 'v1':
bvl = candidatejet.btagDDBvL
cvl = candidatejet.btagDDCvL
cvb = candidatejet.btagDDCvB
elif self._tagger == 'v2':
bvl = candidatejet.btagDDBvLV2
cvl = candidatejet.btagDDCvLV2
cvb = candidatejet.btagDDCvBV2
elif self._tagger == 'v3':
bvl = candidatejet.particleNetMD_Xbb
cvl = candidatejet.particleNetMD_Xcc / (
1 - candidatejet.particleNetMD_Xbb)
cvb = candidatejet.particleNetMD_Xcc / (
candidatejet.particleNetMD_Xcc +
candidatejet.particleNetMD_Xbb)
elif self._tagger == 'v4':
bvl = candidatejet.particleNetMD_Xbb
cvl = candidatejet.btagDDCvLV2
cvb = candidatejet.particleNetMD_Xcc / (
candidatejet.particleNetMD_Xcc +
candidatejet.particleNetMD_Xbb)
else:
raise ValueError("Not an option")
selection.add('minjetkin', (candidatejet.pt >= 450)
& (candidatejet.pt < 1200)
& (candidatejet.msdcorr >= 40.)
& (candidatejet.msdcorr < 201.)
& (abs(candidatejet.eta) < 2.5))
selection.add('_strict_mass', (candidatejet.msdcorr > 85) &
(candidatejet.msdcorr < 130))
selection.add('_high_score', cvl > 0.8)
selection.add('minjetkinmu', (candidatejet.pt >= 400)
& (candidatejet.pt < 1200)
& (candidatejet.msdcorr >= 40.)
& (candidatejet.msdcorr < 201.)
& (abs(candidatejet.eta) < 2.5))
selection.add('minjetkinw', (candidatejet.pt >= 200)
& (candidatejet.pt < 1200)
& (candidatejet.msdcorr >= 40.)
& (candidatejet.msdcorr < 201.)
& (abs(candidatejet.eta) < 2.5))
selection.add('jetid', candidatejet.isTight)
selection.add('n2ddt', (candidatejet.n2ddt < 0.))
if not self._tagger == 'v2':
selection.add('ddbpass', (bvl >= 0.89))
selection.add('ddcpass', (cvl >= 0.83))
selection.add('ddcvbpass', (cvb >= 0.2))
else:
selection.add('ddbpass', (bvl >= 0.7))
selection.add('ddcpass', (cvl >= 0.45))
selection.add('ddcvbpass', (cvb >= 0.03))
jets = events.Jet
jets = jets[(jets.pt > 30.) & (abs(jets.eta) < 2.5) & jets.isTight]
# only consider first 4 jets to be consistent with old framework
jets = jets[:, :4]
dphi = abs(jets.delta_phi(candidatejet))
selection.add(
'antiak4btagMediumOppHem',
ak.max(jets[dphi > np.pi / 2][self._ak4tagBranch],
axis=1,
mask_identity=False) <
BTagEfficiency.btagWPs[self._ak4tagger][self._year]['medium'])
ak4_away = jets[dphi > 0.8]
selection.add(
'ak4btagMedium08',
ak.max(ak4_away[self._ak4tagBranch], axis=1, mask_identity=False) >
BTagEfficiency.btagWPs[self._ak4tagger][self._year]['medium'])
met = events.MET
selection.add('met', met.pt < 140.)
goodmuon = ((events.Muon.pt > 10)
& (abs(events.Muon.eta) < 2.4)
& (events.Muon.pfRelIso04_all < 0.25)
& events.Muon.looseId)
nmuons = ak.sum(goodmuon, axis=1)
leadingmuon = ak.firsts(events.Muon[goodmuon])
if self._looseTau:
goodelectron = ((events.Electron.pt > 10)
& (abs(events.Electron.eta) < 2.5)
&
(events.Electron.cutBased >= events.Electron.VETO))
nelectrons = ak.sum(goodelectron, axis=1)
ntaus = ak.sum(
((events.Tau.pt > 20)
& (abs(events.Tau.eta) < 2.3)
& events.Tau.idDecayMode
& ((events.Tau.idMVAoldDM2017v2 & 2) != 0)
& ak.all(events.Tau.metric_table(events.Muon[goodmuon]) > 0.4,
axis=2)
& ak.all(events.Tau.metric_table(
events.Electron[goodelectron]) > 0.4,
axis=2)),
axis=1,
)
else:
goodelectron = (
(events.Electron.pt > 10)
& (abs(events.Electron.eta) < 2.5)
& (events.Electron.cutBased >= events.Electron.LOOSE))
nelectrons = ak.sum(goodelectron, axis=1)
ntaus = ak.sum(
(events.Tau.pt > 20)
&
events.Tau.idDecayMode # bacon iso looser than Nano selection
& ak.all(events.Tau.metric_table(events.Muon[goodmuon]) > 0.4,
axis=2)
& ak.all(events.Tau.metric_table(events.Electron[goodelectron])
> 0.4,
axis=2),
axis=1,
)
selection.add('noleptons',
(nmuons == 0) & (nelectrons == 0) & (ntaus == 0))
selection.add('onemuon',
(nmuons == 1) & (nelectrons == 0) & (ntaus == 0))
selection.add('muonkin',
(leadingmuon.pt > 55.) & (abs(leadingmuon.eta) < 2.1))
selection.add('muonDphiAK8',
abs(leadingmuon.delta_phi(candidatejet)) > 2 * np.pi / 3)
# W-Tag (Tag and Probe)
# tag side
selection.add(
'ak4btagMediumOppHem',
ak.max(jets[dphi > np.pi / 2][self._ak4tagBranch],
axis=1,
mask_identity=False) >
BTagEfficiency.btagWPs[self._ak4tagger][self._year]['medium'])
selection.add('met40p', met.pt > 40.)
selection.add('tightMuon',
(leadingmuon.tightId) & (leadingmuon.pt > 53.))
# selection.add('ptrecoW', (leadingmuon + met).pt > 250.)
selection.add('ptrecoW200', (leadingmuon + met).pt > 200.)
selection.add(
'ak4btagNearMu',
leadingmuon.delta_r(leadingmuon.nearest(ak4_away, axis=None)) <
2.0)
_bjets = jets[self._ak4tagBranch] > BTagEfficiency.btagWPs[
self._ak4tagger][self._year]['medium']
# _nearAK8 = jets.delta_r(candidatejet) < 0.8
# _nearMu = jets.delta_r(ak.firsts(events.Muon)) < 0.3
# selection.add('ak4btagOld', ak.sum(_bjets & ~_nearAK8 & ~_nearMu, axis=1) >= 1)
_nearAK8 = jets.delta_r(candidatejet) < 0.8
_nearMu = jets.delta_r(leadingmuon) < 0.3
selection.add('ak4btagOld',
ak.sum(_bjets & ~_nearAK8 & ~_nearMu, axis=1) >= 1)
# _nearAK8 = jets.delta_r(candidatejet) < 0.8
# _nearMu = jets.delta_r(candidatejet.nearest(events.Muon[goodmuon], axis=None)) < 0.3
# selection.add('ak4btagNew', ak.sum(_bjets & ~_nearAK8 & ~_nearMu, axis=1) >= 1)
# probe side
selection.add('minWjetpteta',
(candidatejet.pt >= 200) & (abs(candidatejet.eta) < 2.4))
# selection.add('noNearMuon', candidatejet.delta_r(candidatejet.nearest(events.Muon[goodmuon], axis=None)) > 1.0)
selection.add('noNearMuon', candidatejet.delta_r(leadingmuon) > 1.0)
#####
if isRealData:
genflavor = ak.zeros_like(candidatejet.pt)
else:
if 'HToCC' in dataset or 'HToBB' in dataset:
if self._ewkHcorr:
add_HiggsEW_kFactors(weights, events.GenPart, dataset)
weights.add('genweight', events.genWeight)
if "PSWeight" in events.fields:
add_ps_weight(weights, events.PSWeight)
else:
add_ps_weight(weights, None)
if "LHEPdfWeight" in events.fields:
add_pdf_weight(weights, events.LHEPdfWeight)
else:
add_pdf_weight(weights, None)
if "LHEScaleWeight" in events.fields:
add_scalevar_7pt(weights, events.LHEScaleWeight)
add_scalevar_3pt(weights, events.LHEScaleWeight)
else:
add_scalevar_7pt(weights, [])
add_scalevar_3pt(weights, [])
add_pileup_weight(weights, events.Pileup.nPU, self._year, dataset)
bosons = getBosons(events.GenPart)
matchedBoson = candidatejet.nearest(bosons,
axis=None,
threshold=0.8)
if self._tightMatch:
match_mask = (
(candidatejet.pt - matchedBoson.pt) / matchedBoson.pt <
0.5) & ((candidatejet.msdcorr - matchedBoson.mass) /
matchedBoson.mass < 0.3)
selmatchedBoson = ak.mask(matchedBoson, match_mask)
genflavor = bosonFlavor(selmatchedBoson)
else:
genflavor = bosonFlavor(matchedBoson)
genBosonPt = ak.fill_none(ak.firsts(bosons.pt), 0)
if self._newVjetsKfactor:
add_VJets_kFactors(weights, events.GenPart, dataset)
else:
add_VJets_NLOkFactor(weights, genBosonPt, self._year, dataset)
if shift_name is None:
output['btagWeight'].fill(val=self._btagSF.addBtagWeight(
weights, ak4_away, self._ak4tagBranch))
if self._nnlops_rew and dataset in [
'GluGluHToCC_M125_13TeV_powheg_pythia8'
]:
weights.add('minlo_rew',
powheg_to_nnlops(ak.to_numpy(genBosonPt)))
if self._newTrigger:
add_jetTriggerSF(
weights, ak.firsts(fatjets),
self._year if not self._skipRunB else f'{self._year}CDEF',
selection)
else:
add_jetTriggerWeight(weights, candidatejet.msdcorr,
candidatejet.pt, self._year)
add_mutriggerSF(weights, leadingmuon, self._year, selection)
add_mucorrectionsSF(weights, leadingmuon, self._year, selection)
if self._year in ("2016", "2017"):
weights.add("L1Prefiring", events.L1PreFiringWeight.Nom,
events.L1PreFiringWeight.Up,
events.L1PreFiringWeight.Dn)
logger.debug("Weight statistics: %r" % weights.weightStatistics)
msd_matched = candidatejet.msdcorr * self._msdSF[self._year] * (
genflavor > 0) + candidatejet.msdcorr * (genflavor == 0)
regions = {
'signal': [
'noleptons', 'minjetkin', 'met', 'metfilter', 'jetid',
'antiak4btagMediumOppHem', 'n2ddt', 'trigger', 'lumimask'
],
'signal_noddt': [
'noleptons', 'minjetkin', 'met', 'jetid',
'antiak4btagMediumOppHem', 'trigger', 'lumimask', 'metfilter'
],
# 'muoncontrol': ['minjetkinmu', 'jetid', 'n2ddt', 'ak4btagMedium08', 'onemuon', 'muonkin', 'muonDphiAK8', 'muontrigger', 'lumimask', 'metfilter'],
'muoncontrol': [
'onemuon', 'muonkin', 'muonDphiAK8', 'metfilter',
'minjetkinmu', 'jetid', 'ak4btagMedium08', 'n2ddt',
'muontrigger', 'lumimask'
],
'muoncontrol_noddt': [
'onemuon', 'muonkin', 'muonDphiAK8', 'jetid', 'metfilter',
'minjetkinmu', 'jetid', 'ak4btagMedium08', 'muontrigger',
'lumimask'
],
'wtag': [
'onemuon', 'tightMuon', 'minjetkinw', 'jetid', 'met40p',
'metfilter', 'ptrecoW200', 'ak4btagOld', 'muontrigger',
'lumimask'
],
'wtag0': [
'onemuon', 'tightMuon', 'met40p', 'metfilter', 'ptrecoW200',
'ak4btagOld', 'muontrigger', 'lumimask'
],
'wtag2': [
'onemuon', 'tightMuon', 'minjetkinw', 'jetid',
'ak4btagMediumOppHem', 'met40p', 'metfilter', 'ptrecoW200',
'ak4btagOld', 'muontrigger', 'lumimask'
],
'noselection': [],
}
def normalize(val, cut):
if cut is None:
ar = ak.to_numpy(ak.fill_none(val, np.nan))
return ar
else:
ar = ak.to_numpy(ak.fill_none(val[cut], np.nan))
return ar
import time
tic = time.time()
if shift_name is None:
for region, cuts in regions.items():
allcuts = set([])
cut = selection.all(*allcuts)
output['cutflow_msd'].fill(region=region,
genflavor=normalize(
genflavor, None),
cut=0,
weight=weights.weight(),
msd=normalize(msd_matched, None))
output['cutflow_eta'].fill(region=region,
genflavor=normalize(genflavor, cut),
cut=0,
weight=weights.weight()[cut],
eta=normalize(
candidatejet.eta, cut))
output['cutflow_pt'].fill(region=region,
genflavor=normalize(genflavor, cut),
cut=0,
weight=weights.weight()[cut],
pt=normalize(candidatejet.pt, cut))
for i, cut in enumerate(cuts + ['ddcvbpass', 'ddcpass']):
allcuts.add(cut)
cut = selection.all(*allcuts)
output['cutflow_msd'].fill(region=region,
genflavor=normalize(
genflavor, cut),
cut=i + 1,
weight=weights.weight()[cut],
msd=normalize(msd_matched, cut))
output['cutflow_eta'].fill(
region=region,
genflavor=normalize(genflavor, cut),
cut=i + 1,
weight=weights.weight()[cut],
eta=normalize(candidatejet.eta, cut))
output['cutflow_pt'].fill(
region=region,
genflavor=normalize(genflavor, cut),
cut=i + 1,
weight=weights.weight()[cut],
pt=normalize(candidatejet.pt, cut))
if self._evtVizInfo and 'ddcpass' in allcuts and isRealData and region == 'signal':
if 'event' not in events.fields:
continue
_cut = selection.all(*allcuts, '_strict_mass',
'_high_score')
# _cut = selection.all('_strict_mass'')
output['to_check'][
'mass'] += processor.column_accumulator(
normalize(msd_matched, _cut))
nfatjet = ak.sum(
((fatjets.pt > 200) &
(abs(fatjets.eta) < 2.5) & fatjets.isTight),
axis=1)
output['to_check'][
'njet'] += processor.column_accumulator(
normalize(nfatjet, _cut))
output['to_check'][
'fname'] += processor.column_accumulator(
np.array([events.metadata['filename']] *
len(normalize(msd_matched, _cut))))
output['to_check'][
'event'] += processor.column_accumulator(
normalize(events.event, _cut))
output['to_check'][
'luminosityBlock'] += processor.column_accumulator(
normalize(events.luminosityBlock, _cut))
output['to_check'][
'run'] += processor.column_accumulator(
normalize(events.run, _cut))
if shift_name is None:
systematics = [None] + list(weights.variations)
else:
systematics = [shift_name]
def fill(region, systematic, wmod=None):
selections = regions[region]
cut = selection.all(*selections)
sname = 'nominal' if systematic is None else systematic
if wmod is None:
if systematic in weights.variations:
weight = weights.weight(modifier=systematic)[cut]
else:
weight = weights.weight()[cut]
else:
weight = weights.weight()[cut] * wmod[cut]
output['templates'].fill(
region=region,
systematic=sname,
runid=runmap(events.run)[cut],
genflavor=normalize(genflavor, cut),
pt=normalize(candidatejet.pt, cut),
msd=normalize(msd_matched, cut),
ddb=normalize(bvl, cut),
ddc=normalize(cvl, cut),
ddcvb=normalize(cvb, cut),
weight=weight,
)
if region in [
'wtag', 'wtag0', 'wtag2', 'wtag3', 'wtag4', 'wtag5',
'wtag6', 'wtag7', 'noselection'
]: # and sname in ['nominal', 'pileup_weightDown', 'pileup_weightUp', 'jet_triggerDown', 'jet_triggerUp']:
output['wtag'].fill(
region=region,
systematic=sname,
genflavor=normalize(genflavor, cut),
pt=normalize(candidatejet.pt, cut),
msd=normalize(msd_matched, cut),
n2ddt=normalize(candidatejet.n2ddt, cut),
ddc=normalize(cvl, cut),
ddcvb=normalize(cvb, cut),
weight=weight,
)
# if region in ['signal', 'noselection']:
# output['etaphi'].fill(
# region=region,
# systematic=sname,
# runid=runmap(events.run)[cut],
# genflavor=normalize(genflavor, cut),
# pt=normalize(candidatejet.pt, cut),
# eta=normalize(candidatejet.eta, cut),
# phi=normalize(candidatejet.phi, cut),
# ddc=normalize(cvl, cut),
# ddcvb=normalize(cvb, cut),
# ),
if not isRealData:
if wmod is not None:
_custom_weight = events.genWeight[cut] * wmod[cut]
else:
_custom_weight = np.ones_like(weight)
output['genresponse_noweight'].fill(
region=region,
systematic=sname,
pt=normalize(candidatejet.pt, cut),
genpt=normalize(genBosonPt, cut),
weight=_custom_weight,
)
output['genresponse'].fill(
region=region,
systematic=sname,
pt=normalize(candidatejet.pt, cut),
genpt=normalize(genBosonPt, cut),
weight=weight,
)
if systematic is None:
output['signal_opt'].fill(
region=region,
genflavor=normalize(genflavor, cut),
ddc=normalize(cvl, cut),
ddcvb=normalize(cvb, cut),
msd=normalize(msd_matched, cut),
weight=weight,
)
output['signal_optb'].fill(
region=region,
genflavor=normalize(genflavor, cut),
ddb=normalize(bvl, cut),
msd=normalize(msd_matched, cut),
weight=weight,
)
for region in regions:
cut = selection.all(*(set(regions[region]) - {'n2ddt'}))
if shift_name is None:
output['nminus1_n2ddt'].fill(
region=region,
n2ddt=normalize(candidatejet.n2ddt, cut),
weight=weights.weight()[cut],
)
for systematic in systematics:
if isRealData and systematic is not None:
continue
fill(region, systematic)
if shift_name is None and 'GluGluH' in dataset and 'LHEWeight' in events.fields:
for i in range(9):
fill(region, 'LHEScale_%d' % i, events.LHEScaleWeight[:,
i])
for c in events.LHEWeight.fields[1:]:
fill(region, 'LHEWeight_%s' % c, events.LHEWeight[c])
toc = time.time()
output["filltime"] = toc - tic
if shift_name is None:
output["weightStats"] = weights.weightStatistics
return {dataset: output}
def process(self, events):
def normalize(val, cut):
return ak.to_numpy(ak.fill_none(
val[cut],
np.nan)) #val[cut].pad(1, clip=True).fillna(0).flatten()
def fill(region, cuts, systematic=None, wmod=None):
print('filling %s' % region)
selections = cuts
cut = selection.all(*selections)
if 'signal' in region: weight = weights_signal.weight()[cut]
elif 'muonCR' in region: weight = weights_muonCR.weight()[cut]
elif 'VtaggingCR' in region:
weight = weights_VtaggingCR.weight()[cut]
output['templates'].fill(
dataset=dataset,
region=region,
pt=normalize(candidatejet.pt, cut),
msd=normalize(candidatejet.msdcorr, cut),
n2ddt=normalize(candidatejet.n2ddt, cut),
#gruddt=normalize(candidatejet.gruddt, cut),
in_v3_ddt=normalize(candidatejet.in_v3_ddt, cut),
hadW=normalize(candidatejet.nmatcheddau, cut),
weight=weight,
),
output['event'].fill(
dataset=dataset,
region=region,
MET=events.MET.pt[cut],
#nJet=fatjets.counts[cut],
nPFConstituents=normalize(candidatejet.nPFConstituents, cut),
weight=weight,
),
output['deepAK8'].fill(
dataset=dataset,
region=region,
deepTagMDWqq=normalize(candidatejet.deepTagMDWqq, cut),
deepTagMDZqq=normalize(candidatejet.deepTagMDZqq, cut),
msd=normalize(candidatejet.msdcorr, cut),
#genflavor=genflavor[cut],
weight=weight,
),
output['in_v3'].fill(
dataset=dataset,
region=region,
#genflavor=genflavor[cut],
in_v3=normalize(candidatejet.in_v3, cut),
n2=normalize(candidatejet.n2b1, cut),
gru=normalize(candidatejet.gru, cut),
weight=weight,
),
if 'muonCR' in dataset or 'VtaggingCR' in dataset:
output['muon'].fill(
dataset=dataset,
region=region,
mu_pt=normalize(candidatemuon.pt, cut),
mu_eta=normalize(candidatemuon.eta, cut),
mu_pfRelIso04_all=normalize(candidatemuon.pfRelIso04_all,
cut),
weight=weight,
),
#common jet kinematics
gru = events.GRU
IN = events.IN
fatjets = events.FatJet
fatjets['msdcorr'] = corrected_msoftdrop(fatjets)
fatjets['qcdrho'] = 2 * np.log(fatjets.msdcorr / fatjets.pt)
fatjets['gruddt'] = gru.v25 - shift(
fatjets, algo='gruddt', year='2017')
fatjets['gru'] = gru.v25
fatjets['in_v3'] = IN.v3
fatjets['in_v3_ddt'] = IN.v3 - shift(
fatjets, algo='inddt', year='2017')
fatjets['in_v3_ddt_90pctl'] = IN.v3 - shift(
fatjets, algo='inddt90pctl', year='2017')
fatjets['n2ddt'] = fatjets.n2b1 - n2ddt_shift(fatjets, year='2017')
fatjets['nmatcheddau'] = TTsemileptonicmatch(events)
dataset = events.metadata['dataset']
print('process dataset', dataset)
isRealData = not hasattr(events, 'genWeight')
output = self.accumulator.identity()
if (len(events) == 0): return output
selection = PackedSelection('uint64')
weights_signal = Weights(len(events))
weights_muonCR = Weights(len(events))
weights_VtaggingCR = Weights(len(events))
if not isRealData:
output['sumw'][dataset] += ak.sum(events.genWeight)
#######################
if 'signal' in self._region:
if isRealData:
trigger_fatjet = np.zeros(len(events), dtype='bool')
for t in self._triggers[self._year]:
try:
trigger_fatjet = trigger_fatjet | events.HLT[t]
except:
print('trigger %s not available' % t)
continue
else:
trigger_fatjet = np.ones(len(events), dtype='bool')
fatjets["genMatchFull"] = VQQgenmatch(events)
candidatejet = ak.firsts(fatjets)
candidatejet["genMatchFull"] = VQQgenmatch(events)
nelectrons = ak.sum(
(events.Electron.pt > 10.)
& (abs(events.Electron.eta) < 2.5)
& (events.Electron.cutBased >= events.Electron.VETO),
axis=1,
)
nmuons = ak.sum(
(events.Muon.pt > 10)
& (abs(events.Muon.eta) < 2.1)
& (events.Muon.pfRelIso04_all < 0.4)
& (events.Muon.looseId),
axis=1,
)
ntaus = ak.sum(
(events.Tau.pt > 20.)
& (events.Tau.idDecayMode)
& (events.Tau.rawIso < 5)
& (abs(events.Tau.eta) < 2.3),
axis=1,
)
cuts = {
"S_fatjet_trigger":
trigger_fatjet,
"S_pt":
candidatejet.pt > 525,
"S_eta": (abs(candidatejet.eta) < 2.5),
"S_msdcorr": (candidatejet.msdcorr > 40),
"S_rho":
((candidatejet.qcdrho > -5.5) & (candidatejet.qcdrho < -2.)),
"S_jetid": (candidatejet.isTight),
"S_VQQgenmatch": (candidatejet.genMatchFull),
"S_noelectron": (nelectrons == 0),
"S_nomuon": (nmuons == 0),
"S_notau": (ntaus == 0),
}
for name, cut in cuts.items():
print(name, cut)
selection.add(name, cut)
if isRealData:
genflavor = 0 #candidatejet.pt.zeros_like().pad(1, clip=True).fillna(-1).flatten()
if not isRealData:
weights_signal.add('genweight', events.genWeight)
#add_pileup_weight(weights_signal, events.Pileup.nPU, self._year, dataset)
add_jetTriggerWeight(weights_signal, candidatejet.msdcorr,
candidatejet.pt, self._year)
bosons = getBosons(events.GenPart)
genBosonPt = ak.fill_none(ak.firsts(bosons.pt), 0)
add_VJets_NLOkFactor(weights_signal, genBosonPt, self._year,
dataset)
#genflavor = matchedBosonFlavor(candidatejet, bosons).pad(1, clip=True).fillna(-1).flatten()
allcuts_signal = set()
output['cutflow_signal'][dataset]['none'] += float(
weights_signal.weight().sum())
for cut in cuts:
allcuts_signal.add(cut)
output['cutflow_signal'][dataset][cut] += float(
weights_signal.weight()[selection.all(
*allcuts_signal)].sum())
fill('signal', cuts.keys())
#######################
if 'muonCR' in self._region:
if isRealData:
trigger_muon = np.zeros(len(events), dtype='bool')
for t in self._muontriggers[self._year]:
trigger_muon = trigger_muon | events.HLT[t]
else:
trigger_muon = np.ones(len(events), dtype='bool')
candidatejet = ak.firsts(fatjets)
candidatemuon = events.Muon[:, :5]
jets = events.Jet[((events.Jet.pt > 50.)
& (abs(events.Jet.eta) < 2.5)
& (events.Jet.isTight))][:, :4]
dphi = abs(jets.delta_phi(candidatejet))
ak4_away = jets[(dphi > 0.8)]
nelectrons = ak.sum(
(events.Electron.pt > 10.)
& (abs(events.Electron.eta) < 2.5)
& (events.Electron.cutBased >= events.Electron.VETO),
axis=1,
)
nmuons = ak.sum(
(events.Muon.pt > 10)
& (abs(events.Muon.eta) < 2.4)
& (events.Muon.pfRelIso04_all < 0.25)
& (events.Muon.looseId),
axis=1,
)
ntaus = ak.sum(
(events.Tau.pt > 20.)
& (events.Tau.idDecayMode)
& (events.Tau.rawIso < 5)
& (abs(events.Tau.eta) < 2.3)
& (events.Tau.idMVAoldDM2017v1 >= 16),
axis=1,
)
cuts = {
"CR1_muon_trigger":
trigger_muon,
"CR1_jet_pt": (candidatejet.pt > 525),
"CR1_jet_eta": (abs(candidatejet.eta) < 2.5),
"CR1_jet_msd": (candidatejet.msdcorr > 40),
"CR1_jet_rho":
((candidatejet.qcdrho > -5.5) & (candidatejet.qcdrho < -2.)),
"CR1_mu_pt":
ak.any(candidatemuon.pt > 55, axis=1),
"CR1_mu_eta":
ak.any(abs(candidatemuon.eta) < 2.1, axis=1),
"CR1_mu_IDLoose":
ak.any(candidatemuon.looseId, axis=1),
"CR1_mu_isolationTight":
ak.any(candidatemuon.pfRelIso04_all < 0.15, axis=1),
"CR1_muonDphiAK8":
ak.any(
abs(candidatemuon.delta_phi(candidatejet)) > 2 * np.pi / 3,
axis=1),
"CR1_ak4btagMedium08":
(ak.max(ak4_away.btagCSVV2, axis=1, mask_identity=False) >
BTagEfficiency.btagWPs[self._year]['medium']
), #(ak4_away.btagCSVV2.max() > 0.8838),
"CR1_noelectron": (nelectrons == 0),
"CR1_onemuon": (nmuons == 1),
"CR1_notau": (ntaus == 0),
}
for name, cut in cuts.items():
selection.add(name, cut)
if isRealData:
genflavor = 0 #candidatejet.pt.zeros_like().pad(1, clip=True).fillna(-1).flatten()
if not isRealData:
weights_muonCR.add('genweight', events.genWeight)
#add_pileup_weight(weights_muonCR, events.Pileup.nPU, self._year, dataset)
#add_singleMuTriggerWeight(weights, candidatejet.msdcorr, candidatejet.pt, self._year)
bosons = getBosons(events.GenPart)
genBosonPt = ak.fill_none(ak.firsts(bosons.pt), 0)
#add_VJets_NLOkFactor(weights, genBosonPt, self._year, dataset)
#genflavor = matchedBosonFlavor(candidatejet, bosons).pad(1, clip=True).fillna(-1).flatten()
allcuts_ttbar_muoncontrol = set()
output['cutflow_muonCR'][dataset]['none'] += float(
weights_muonCR.weight().sum())
for cut in cuts:
allcuts_ttbar_muoncontrol.add(cut)
output['cutflow_muonCR'][dataset][cut] += float(
weights_muonCR.weight()[selection.all(
*allcuts_ttbar_muoncontrol)].sum())
fill('muonCR', cuts.keys())
#######################
if 'VtaggingCR' in self._region:
if isRealData:
trigger_muon = np.zeros(len(events), dtype='bool')
for t in self._muontriggers[self._year]:
trigger_muon = trigger_muon | events.HLT[t]
else:
trigger_muon = np.ones(len(events), dtype='bool')
candidatejet = ak.firsts(fatjets)
candidatemuon = ak.firsts(events.Muon)
jets = events.Jet[((events.Jet.pt > 30.)
& (abs(events.Jet.eta) < 2.4))][:, :4]
dr_ak4_ak8 = jets.delta_r(candidatejet)
dr_ak4_muon = jets.delta_r(candidatemuon)
ak4_away = jets[(dr_ak4_ak8 > 0.8)] # & (dr_ak4_muon > 0.4)]
mu_p4 = ak.zip(
{
"pt": ak.fill_none(candidatemuon.pt, 0),
"eta": ak.fill_none(candidatemuon.eta, 0),
"phi": ak.fill_none(candidatemuon.phi, 0),
"mass": ak.fill_none(candidatemuon.mass, 0),
},
with_name="PtEtaPhiMLorentzVector")
met_p4 = ak.zip(
{
"pt": ak.from_iter([[v] for v in events.MET.pt]),
"eta": ak.from_iter([[v] for v in np.zeros(len(events))]),
"phi": ak.from_iter([[v] for v in events.MET.phi]),
"mass": ak.from_iter([[v] for v in np.zeros(len(events))]),
},
with_name="PtEtaPhiMLorentzVector")
Wleptoniccandidate = mu_p4 + met_p4
nelectrons = ak.sum(
((events.Electron.pt > 10.)
& (abs(events.Electron.eta) < 2.5)
& (events.Electron.cutBased >= events.Electron.VETO)),
axis=1,
)
n_tight_muon = ak.sum(
((events.Muon.pt > 53)
& (abs(events.Muon.eta) < 2.1)
& (events.Muon.tightId)),
axis=1,
)
n_loose_muon = ak.sum(
((events.Muon.pt > 20)
& (events.Muon.looseId)
& (abs(events.Muon.eta) < 2.4)),
axis=1,
)
ntaus = ak.sum(
((events.Tau.pt > 20.)
& (events.Tau.idDecayMode)
& (events.Tau.rawIso < 5)
& (abs(events.Tau.eta) < 2.3)
& (events.Tau.idMVAoldDM2017v1 >= 16)),
axis=1,
)
cuts = {
"CR2_muon_trigger":
trigger_muon,
"CR2_jet_pt": (candidatejet.pt > 200),
"CR2_jet_eta": (abs(candidatejet.eta) < 2.5),
"CR2_jet_msd": (candidatejet.msdcorr > 40),
"CR2_mu_pt":
candidatemuon.pt > 53,
"CR2_mu_eta": (abs(candidatemuon.eta) < 2.1),
"CR2_mu_IDTight":
candidatemuon.tightId,
"CR2_mu_isolationTight": (candidatemuon.pfRelIso04_all < 0.15),
"CR2_muonDphiAK8":
abs(candidatemuon.delta_phi(candidatejet)) > 2 * np.pi / 3,
"CR2_ak4btagMedium08":
(ak.max(ak4_away.btagCSVV2, axis=1, mask_identity=False) >
BTagEfficiency.btagWPs[self._year]['medium']),
"CR2_leptonicW":
ak.flatten(Wleptoniccandidate.pt > 200),
"CR2_MET": (events.MET.pt > 40.),
"CR2_noelectron": (nelectrons == 0),
"CR2_one_tightMuon": (n_tight_muon == 1),
"CR2_one_looseMuon": (n_loose_muon == 1),
#"CR2_notau" : (ntaus==0),
}
for name, cut in cuts.items():
print(name, cut)
selection.add(name, cut)
#weights.add('metfilter', events.Flag.METFilters)
if isRealData:
genflavor = 0 #candidatejet.pt.zeros_like().pad(1, clip=True).fillna(-1).flatten()
if not isRealData:
weights_VtaggingCR.add('genweight', events.genWeight)
#add_pileup_weight(weights_VtaggingCR, events.Pileup.nPU, self._year, dataset)
#add_singleMuTriggerWeight(weights, abs(candidatemuon.eta), candidatemuon.pt, self._year)
bosons = getBosons(events.GenPart)
genBosonPt = ak.fill_none(ak.firsts(bosons.pt), 0)
#add_VJets_NLOkFactor(weights, genBosonPt, self._year, dataset)
#genflavor = matchedBosonFlavor(candidatejet, bosons).pad(1, clip=True).fillna(-1).flatten()
#b-tag weights
allcuts_vselection = set()
output['cutflow_VtaggingCR'][dataset]['none'] += float(
weights_VtaggingCR.weight().sum())
for cut in cuts:
allcuts_vselection.add(cut)
output['cutflow_VtaggingCR'][dataset][cut] += float(
weights_VtaggingCR.weight()[selection.all(
*allcuts_vselection)].sum())
fill('VtaggingCR', cuts.keys())
return output