def pad_and_flatten(val):
import awkward as ak
try:
return ak.flatten(ak.fill_none(ak.pad_none(val, 1, clip=True), 0))
#return val.pad(1, clip=True).fillna(0.).flatten()#.reshape(-1, 1)
except ValueError:
return ak.flatten(val)
def embed_crossref(source, idx_name, dest, dest_name):
"""Embed a cross-reference
Parameters
----------
source : ak.Array
any array with shape N * var * {record}
idx_name : str
A field in the source record
dest : ak.Array
any array with shape N * var * {record}, where:
``ak.max(source[idx_name], axis=1) < ak.num(dest)`` and
``ak.min(source[idx_name], axis=1) >= 0``
"""
print(ak.max(source[idx_name], axis=1))
print(ak.num(dest))
print(ak.all(ak.max(source[idx_name], axis=1) < ak.num(dest)))
assert ak.all(ak.max(source[idx_name], axis=1) < ak.num(dest))
assert ak.all(ak.min(source[idx_name], axis=1) >= 0)
id_global = ak.flatten(
source[idx_name] + np.asarray(dest.layout.starts), axis=None
)
source[dest_name] = ak.Array(
ak.layout.ListOffsetArray64(
source.layout.offsets,
ak.layout.ListOffsetArray64(
source.layout.content.offsets,
ak.flatten(dest)[id_global].layout,
),
)
)
def test_ByteMaskedArray_flatten():
content = awkward1.from_iter([[[0.0, 1.1, 2.2], [], [3.3, 4.4]], [], [[5.5]], [[6.6, 7.7, 8.8, 9.9]], [[], [10.0, 11.1, 12.2]]], highlevel=False)
mask = awkward1.layout.Index8(numpy.array([0, 0, 1, 1, 0], dtype=numpy.int8))
array = awkward1.Array(awkward1.layout.ByteMaskedArray(mask, content, valid_when=False))
assert awkward1.to_list(array) == [[[0.0, 1.1, 2.2], [], [3.3, 4.4]], [], None, None, [[], [10.0, 11.1, 12.2]]]
assert awkward1.to_list(awkward1.flatten(array, axis=1)) == [[0.0, 1.1, 2.2], [], [3.3, 4.4], [], [10.0, 11.1, 12.2]]
assert awkward1.to_list(awkward1.flatten(array, axis=2)) == [[0.0, 1.1, 2.2, 3.3, 4.4], [], None, None, [10.0, 11.1, 12.2]]
def yahist_2D_lookup(h, ar1, ar2):
'''
takes a yahist 2D histogram (which has a lookup function) and an awkward array.
'''
return ak.unflatten(
h.lookup(
ak.to_numpy(ak.flatten(ar1)),
ak.to_numpy(ak.flatten(ar2)),
), ak.num(ar1) )
def process(self, events):
output = self.accumulator.identity()
dataset = events.metadata["dataset"]
dimuon = ak.combinations(events.Muon, 2)
dimuon = dimuon["0"] + dimuon["1"]
output["pt"].fill(dataset=dataset, pt=ak.flatten(events.Muon.pt))
output["mass"].fill(dataset=dataset, mass=ak.flatten(dimuon.mass))
output["cutflow"]["%s_pt" % dataset] += sum(ak.num(events.Muon))
output["cutflow"]["%s_mass" % dataset] += sum(ak.num(dimuon))
return output
def test_flatten_RecordArray():
array = awkward1.Array([{
"x": [],
"y": [[3, 3, 3]]
}, {
"x": [[1]],
"y": [[2, 2]]
}, {
"x": [[2], [2]],
"y": [[1]]
}, {
"x": [[3], [3], [3]],
"y": [[]]
}])
assert awkward1.to_list(awkward1.flatten(array, axis=2)) == [{
"x": [],
"y": [3, 3, 3]
}, {
"x": [1],
"y": [2, 2]
}, {
"x": [2, 2],
"y": [1]
}, {
"x": [3, 3, 3],
"y": []
}]
assert awkward1.to_list(awkward1.flatten(array[1:], axis=2)) == [{
"x": [1],
"y": [2, 2]
}, {
"x": [2, 2],
"y": [1]
}, {
"x": [3, 3, 3],
"y": []
}]
assert awkward1.to_list(awkward1.flatten(array[:, 1:], axis=2)) == [{
"x": [],
"y": []
}, {
"x": [],
"y": []
}, {
"x": [2],
"y": []
}, {
"x": [3, 3],
"y": []
}]
def test_flatten_UnionArray():
content1 = awkward1.from_iter([[1.1], [2.2, 2.2], [3.3, 3.3, 3.3]],
highlevel=False)
content2 = awkward1.from_iter(
[[[3, 3, 3], [3, 3, 3], [3, 3, 3]], [[2, 2], [2, 2]], [[1]]],
highlevel=False)
tags = awkward1.layout.Index8(
numpy.array([0, 1, 0, 1, 0, 1], dtype=numpy.int8))
index = awkward1.layout.Index64(
numpy.array([0, 0, 1, 1, 2, 2], dtype=numpy.int64))
array = awkward1.Array(
awkward1.layout.UnionArray8_64(tags, index, [content1, content2]))
assert awkward1.to_list(array) == [[1.1], [[3, 3, 3], [3, 3, 3], [3, 3,
3]],
[2.2, 2.2], [[2, 2], [2, 2]],
[3.3, 3.3, 3.3], [[1]]]
assert awkward1.to_list(array[1:]) == [[[3, 3, 3], [3, 3, 3], [3, 3, 3]],
[2.2, 2.2], [[2, 2], [2, 2]],
[3.3, 3.3, 3.3], [[1]]]
assert awkward1.to_list(awkward1.flatten(array)) == [
1.1, [3, 3, 3], [3, 3, 3], [3, 3, 3], 2.2, 2.2, [2, 2], [2, 2], 3.3,
3.3, 3.3, [1]
]
assert awkward1.to_list(awkward1.flatten(array[1:])) == [[3, 3, 3],
[3, 3, 3],
[3, 3, 3],
2.2, 2.2, [2, 2],
[2, 2], 3.3, 3.3,
3.3, [1]]
array = awkward1.Array(
awkward1.layout.UnionArray8_64(tags, index, [content2, content2]))
assert awkward1.to_list(array) == [[[3, 3, 3], [3, 3, 3], [3, 3, 3]],
[[3, 3, 3], [3, 3, 3], [3, 3, 3]],
[[2, 2], [2, 2]], [[2, 2], [2, 2]],
[[1]], [[1]]]
assert awkward1.to_list(awkward1.flatten(
array, axis=2)) == [[3, 3, 3, 3, 3, 3, 3, 3, 3],
[3, 3, 3, 3, 3, 3, 3, 3, 3], [2, 2, 2, 2],
[2, 2, 2, 2], [1], [1]]
assert awkward1.to_list(awkward1.flatten(
array[1:], axis=2)) == [[3, 3, 3, 3, 3, 3, 3, 3, 3], [2, 2, 2, 2],
[2, 2, 2, 2], [1], [1]]
assert awkward1.to_list(awkward1.flatten(array[:, 1:],
axis=2)) == [[3, 3, 3, 3, 3, 3],
[3, 3, 3, 3, 3, 3],
[2, 2], [2, 2], [],
[]]
def get_vec(var, mass):
dict = {}
for key, val in key_dict.items():
dict[key] = ak.flatten(ak.Array([evtDict["HH4V"]['genHiggs1' + var + val][h1W], evtDict["HH4V"]['genHiggs2' + var + val][h2W]]))
dict['mass'] = np.ones(totW) * mass
return ak.zip(dict, with_name="PtEtaPhiMLorentzVector")
def process(self, df):
## objects used for cuts
vars_noCut = utl.varGetter(df, self.dataset, self.scaleFactor)
# Our preselection
cuts = bl.cutList(df, vars_noCut, SVJCut=False)
if self.setupNPArr is None:
self.setupNPArray(cuts, variables)
output = self.accumulator.identity()
# run cut loop
cut = cuts["_npz"]
weight = vars_noCut["evtw"][0][cut]
if len(weight) > 0:
for varName, varDetail in variables.items():
# only store jetAK8 variables
if varDetail[4] == 1:
hIn = vars_noCut[varName][0][cut]
# properly flatten certain inputs
if varDetail[5] == 1:
hIn = hIn.flatten()
elif varDetail[5] == 2:
hIn = ak.flatten(hIn)
output['{}'.format(varName)] += col_accumulator(hIn)
return output
def globalindex():
gidx = awkward1.Array(
awkward1.layout.ListOffsetArray32(
awkward1.layout.Index32(source_offsets), index.generator(),
)
)
gidx = gidx.mask[gidx >= 0] + target_offsets[:-1]
return awkward1.fill_none(awkward1.flatten(gidx), -1)
def doAwkwardLookup(h, ar):
'''
takes a ya_hist histogram (which has a lookup function) and an awkward array.
'''
return ak.unflatten(
h.lookup(
ak.to_numpy(
ak.flatten(ar)
)
), ak.num(ar) )
def debug_eff(llp_accept, csc_accept, emtf_accept, gmt_accept, csc_endcap,
csc_station, csc_ring, csc_chamber, emtf_endcap, emtf_station):
# Create a mask for events in acceptance
llp_pass = np.array(ak.sum(llp_accept, axis=-1) > 0)
# Apppply mask to the csc, emtf, and gmt
csc_pass = csc_accept[llp_pass]
emtf_pass = emtf_accept[llp_pass]
gmt_pass = gmt_accept[llp_pass]
mismatch = ak.num(emtf_pass) > ak.flatten(gmt_pass) * 5
print("Mismatch:")
print(mismatch)
print('')
csc_pass_endcap = csc_endcap[llp_pass]
csc_pass_station = csc_station[llp_pass]
csc_pass_ring = csc_ring[llp_pass]
csc_pass_chamber = csc_chamber[llp_pass]
print("CSC Masked Array:")
print(csc_pass[mismatch])
print("CSC endcap:")
print(csc_pass_endcap[mismatch])
print("CSC station:")
print(csc_pass_station[mismatch])
print("CSC ring:")
print(csc_pass_ring[mismatch])
print("CSC chamber:")
print(csc_pass_chamber[mismatch])
print("")
emtf_pass_endcap = emtf_endcap[llp_pass]
emtf_pass_station = emtf_station[llp_pass]
print("EMTF Masked Array:")
print(emtf_pass[mismatch])
print("EMTF endcap:")
print(emtf_pass_endcap[mismatch])
print("EMTF sector:")
print(emtf_pass_station[mismatch])
print("")
print("GMT Masked Array:")
print(gmt_pass[mismatch])
print("")
# Calculate efficiency for each
n_acc = np.count_nonzero(llp_pass)
csc_eff = np.count_nonzero(ak.num(csc_pass, axis=1)) / n_acc * 100
emtf_eff = np.count_nonzero(ak.num(emtf_pass, axis=1)) / n_acc * 100
gmt_eff = np.count_nonzero(gmt_pass) / n_acc * 100
return [n_acc, csc_eff, emtf_eff, gmt_eff]
def local2global(stack):
"""Turn jagged local index into global index
Signature: index,target_offsets,!local2global
Outputs a content array with same shape as index content
"""
target_offsets = stack.pop()
index = stack.pop()
index = index.mask[index >= 0] + target_offsets[:-1]
out = numpy.array(awkward1.flatten(awkward1.fill_none(index, -1)))
if out.dtype != numpy.int64:
raise RuntimeError
stack.append(out)
def __getitem__(self, key):
if key in self._dict:
return self._dict[key]
elif key in self._tree:
self._materialized.add(key)
array = self._tree[key].array(**self._branchargs)
if self._flatten and isinstance(
awkward1.type(array).type, awkward1.types.ListType):
array = awkward1.flatten(array)
array = awkward1.to_awkward0(array)
self._dict[key] = array
return self._dict[key]
else:
raise KeyError(key)
def test_flatten0():
array = awkward1.Array([1.1, 2.2, None, 3.3, None, None, 4.4, 5.5])
assert awkward1.to_list(awkward1.flatten(
array, axis=0)) == [1.1, 2.2, 3.3, 4.4, 5.5]
content0 = awkward1.from_iter([1.1, 2.2, None, 3.3, None, None, 4.4, 5.5],
highlevel=False)
content1 = awkward1.from_iter(["one", None, "two", None, "three"],
highlevel=False)
array = awkward1.Array(
awkward1.layout.UnionArray8_64(
awkward1.layout.Index8(
numpy.array([0, 1, 0, 1, 0, 1, 0, 1, 0, 0, 0, 1, 0],
dtype=numpy.int8)),
awkward1.layout.Index64(
numpy.array([0, 0, 1, 1, 2, 2, 3, 3, 4, 5, 6, 4, 7],
dtype=numpy.int64)), [content0, content1]))
assert awkward1.to_list(array) == [
1.1, "one", 2.2, None, None, "two", 3.3, None, None, None, 4.4,
"three", 5.5
]
assert awkward1.to_list(awkward1.flatten(
array, axis=0)) == [1.1, "one", 2.2, "two", 3.3, 4.4, "three", 5.5]
def matchJets(self, obj, jet, deltaRCut=0.4):
combs = ak.cartesian([obj, jet], nested=True)
jet_index = ak.local_index(delta_r(
combs['0'], combs['1']))[delta_r(combs['0'], combs['1']) < 0.4]
jet_index_pad = ak.flatten(ak.fill_none(
ak.pad_none(jet_index, target=1, clip=True, axis=2), 0),
axis=2)
mask = ak.num(jet_index,
axis=2) > 0 # a mask for obj with a matched jet
mask_match = mask * 1 + ~mask * 0
mask_nomatch = mask * 0 + ~mask * 1
return jet_index_pad, mask_match, mask_nomatch
def expr_to_vals(expr):
vals = eval(expr, dict(), loc)
# if varexp is a simple constant, broadcast it to an array
if _array_ndim(vals) == 0:
vals = vals * np.ones(len(df))
if sel:
if _array_ndim(vals) < _array_ndim(globalmask):
vals, _ = awkward1.broadcast_arrays(vals, globalmask)
vals = vals[globalmask]
if _array_ndim(vals) > 1:
vals = awkward1.flatten(vals)
vals = awkward1.to_numpy(vals)
return vals
def errorbar_hist(P, var, P_name, title, units, low, high, bins):
fig, ax = plt.subplots(figsize=(8, 8))
bin_w = (high - low) / bins
counts, bin_edges = np.histogram(ak.to_list(ak.flatten(P[var])),
bins,
range=(low, high))
bin_centres = (bin_edges[:-1] + bin_edges[1:]) / 2.
err = np.sqrt(counts)
plt.errorbar(bin_centres, counts, yerr=err, fmt='o', color='k')
plt.xlabel(f"{title} [{units}]", fontsize=30)
plt.ylabel("Candidates / (%.4f %s)" % (bin_w, units), fontsize=30)
plt.xlim(low, high)
ax.tick_params(axis='both', which='major', labelsize=25)
ymin, ymax = plt.ylim()
plt.ylim(0., ymax * 1.1)
plt.tight_layout()
plt.show()
fig.savefig(f"{loc.PLOTS}/{P_name}_{var}.pdf")
def errorbar_hist(P, var, P_name, title, low, high):
fig, ax = plt.subplots(figsize=(8, 8))
#Number of events, use this to determine bins and thus bin width
n = np.sum(ak.num(P))
bins = int(np.sqrt(n))
bin_w = (high - low) / bins
counts, bin_edges = np.histogram(ak.to_list(ak.flatten(P[var])),
bins,
range=(low, high))
bin_centres = (bin_edges[:-1] + bin_edges[1:]) / 2.
err = np.sqrt(counts)
plt.errorbar(bin_centres, counts, yerr=err, fmt='o', color='k')
plt.xlabel(title, fontsize=30)
plt.ylabel("Candidates / (%.4f GeV/$c^2$)" % bin_w, fontsize=30)
plt.xlim(low, high)
ax.tick_params(axis='both', which='major', labelsize=25)
ymin, ymax = plt.ylim()
plt.ylim(0., ymax * 1.1)
plt.tight_layout()
plt.show()
fig.savefig(f"{loc.PLOTS}/{P_name}_{var}.pdf")
genPart = final.GenPart
tops = genPart[abs(genPart.pdgId) == 6]
#The isLastCopy Flag filters out copy Genparticles:
tops = tops[tops.hasFlags('isLastCopy')]
tDecay = tops.distinctChildren
tDecay = tDecay[tDecay.hasFlags('isLastCopy')]
t_Events = tDecay[abs(tDecay.pdgId) == 5]
W = tDecay[abs(tDecay.pdgId) == 24]
W = W[W.hasFlags('isLastCopy')]
WDecay = W.distinctChildren
WDecay = WDecay[WDecay.hasFlags('isLastCopy')]
#t_events is the lone bottom, W_events is the -> two jets
#select the hadronically decaying W
W_Events = ak.flatten(WDecay[ak.all(abs(WDecay.pdgId) <= 8, axis=-1)], axis=3)
#HadW is mask for Quark deacying W boson
hadW = ak.num(W_Events, axis=2) == 2
#filters out t_events that have a hadronically decayign W Boson
hadB = t_Events[hadW]
hadB = ak.flatten(hadB, axis=2)
W_quarks = W_Events[hadW]
W_quarks = ak.flatten(W_quarks, axis=2)
#concatentating these two arrays make an array of events with the correctly decaying GenParticles.
qqb = ak.concatenate([hadB, W_quarks], axis=1)
print("qqb Genparts matched")
#Filtering Out events with extra tops
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", "tightSSTTH").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", "tightSSTTH").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)
# define the weight
weight = Weights( len(ev) )
if not re.search(re.compile('MuonEG|DoubleMuon|DoubleEG|EGamma'), dataset):
# 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))
cutflow = Cutflow(output, ev, weight=weight)
sel = Selection(
dataset = dataset,
events = ev,
year = self.year,
ele = electron,
ele_veto = vetoelectron,
mu = muon,
mu_veto = vetomuon,
jet_all = jet,
jet_central = central,
jet_btag = btag,
jet_fwd = fwd,
met = ev.MET,
)
BL = sel.dilep_baseline(cutflow=cutflow, SS=False)
# first, make a few super inclusive plots
output['PV_npvs'].fill(dataset=dataset, multiplicity=ev.PV[BL].npvs, weight=weight.weight()[BL])
output['PV_npvsGood'].fill(dataset=dataset, multiplicity=ev.PV[BL].npvsGood, weight=weight.weight()[BL])
output['N_jet'].fill(dataset=dataset, multiplicity=ak.num(jet)[BL], weight=weight.weight()[BL])
BL_minusNb = sel.dilep_baseline(SS=False, omit=['N_btag>0'])
output['N_b'].fill(dataset=dataset, multiplicity=ak.num(btag)[BL_minusNb], weight=weight.weight()[BL_minusNb])
output['N_central'].fill(dataset=dataset, multiplicity=ak.num(central)[BL], weight=weight.weight()[BL])
output['N_ele'].fill(dataset=dataset, multiplicity=ak.num(electron)[BL], weight=weight.weight()[BL])
output['N_mu'].fill(dataset=dataset, multiplicity=ak.num(electron)[BL], weight=weight.weight()[BL])
BL_minusFwd = sel.dilep_baseline(SS=False, omit=['N_fwd>0'])
output['N_fwd'].fill(dataset=dataset, multiplicity=ak.num(fwd)[BL_minusFwd], weight=weight.weight()[BL_minusFwd])
BL_minusMET = sel.dilep_baseline(SS=False, omit=['MET>50'])
output['MET'].fill(
dataset = dataset,
pt = ev.MET[BL_minusMET].pt,
phi = ev.MET[BL_minusMET].phi,
weight = weight.weight()[BL_minusMET]
)
#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]
)
if re.search(re.compile('MuonEG|DoubleMuon|DoubleEG|EGamma'), dataset):
#rle = ak.to_numpy(ak.zip([ev.run, ev.luminosityBlock, ev.event]))
run_ = ak.to_numpy(ev.run)
lumi_ = ak.to_numpy(ev.luminosityBlock)
event_ = ak.to_numpy(ev.event)
output['%s_run'%dataset] += processor.column_accumulator(run_[BL])
output['%s_lumi'%dataset] += processor.column_accumulator(lumi_[BL])
output['%s_event'%dataset] += processor.column_accumulator(event_[BL])
# Now, take care of systematic unceratinties
if not re.search(re.compile('MuonEG|DoubleMuon|DoubleEG|EGamma'), dataset):
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]
met = ev.MET
met['pt'] = getattr(met, var)
sel = Selection(
dataset = dataset,
events = ev,
year = self.year,
ele = electron,
ele_veto = vetoelectron,
mu = muon,
mu_veto = vetomuon,
jet_all = jet,
jet_central = central,
jet_btag = btag,
jet_fwd = fwd,
met = met,
)
BL = sel.dilep_baseline(SS=False)
# 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)[BL], weight=weight.weight()[BL])
BL_minusFwd = sel.dilep_baseline(SS=False, omit=['N_fwd>0'])
output['N_fwd_'+var].fill(dataset=dataset, multiplicity=ak.num(fwd)[BL_minusFwd], weight=weight.weight()[BL_minusFwd])
BL_minusNb = sel.dilep_baseline(SS=False, omit=['N_btag>0'])
output['N_b_'+var].fill(dataset=dataset, multiplicity=ak.num(btag)[BL_minusNb], weight=weight.weight()[BL_minusNb])
output['N_central_'+var].fill(dataset=dataset, multiplicity=ak.num(central)[BL], weight=weight.weight()[BL])
# 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),
#p = ak.flatten(high_p_fwd[BL].p),
eta = ak.flatten(high_p_fwd[BL].eta),
phi = ak.flatten(high_p_fwd[BL].phi),
weight = weight.weight()[BL]
)
BL_minusMET = sel.dilep_baseline(SS=False, omit=['MET>50'])
output['MET_'+var].fill(
dataset = dataset,
pt = getattr(ev.MET, var)[BL_minusMET],
phi = ev.MET[BL_minusMET].phi,
weight = weight.weight()[BL_minusMET]
)
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)
if not re.search(re.compile('MuonEG|DoubleMuon|DoubleEG|EGamma'), dataset):
## 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]
dilepton_mass = (leading_lepton+trailing_lepton).mass
dilepton_pt = (leading_lepton+trailing_lepton).pt
dilepton_dR = delta_r(leading_lepton, trailing_lepton)
lepton_pdgId_pt_ordered = ak.fill_none(ak.pad_none(lepton[ak.argsort(lepton.pt, ascending=False)].pdgId, 2, clip=True), 0)
if not re.search(re.compile('MuonEG|DoubleMuon|DoubleEG|EGamma'), dataset):
n_nonprompt = getNonPromptFromFlavour(electron) + getNonPromptFromFlavour(muon)
n_chargeflip = getChargeFlips(electron, ev.GenPart) + getChargeFlips(muon, ev.GenPart)
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)
## Tau and other stuff
tau = getTaus(ev)
track = getIsoTracks(ev)
## 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)
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)
## 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)
# define the weight
weight = Weights( len(ev) )
if not re.search(re.compile('MuonEG|DoubleMuon|DoubleEG|EGamma'), dataset):
# 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))
cutflow = Cutflow(output, ev, weight=weight)
sel = Selection(
dataset = dataset,
events = ev,
year = self.year,
ele = electron,
ele_veto = vetoelectron,
mu = muon,
mu_veto = vetomuon,
jet_all = jet,
jet_central = central,
jet_btag = btag,
jet_fwd = fwd,
met = ev.MET,
)
BL = sel.dilep_baseline(cutflow=cutflow, SS=True)
weight_BL = weight.weight()[BL]
if True:
# define the inputs to the NN
# this is super stupid. there must be a better way.
NN_inputs = np.stack([
ak.to_numpy(ak.num(jet[BL])),
ak.to_numpy(ak.num(tau[BL])),
ak.to_numpy(ak.num(track[BL])),
ak.to_numpy(st[BL]),
ak.to_numpy(ev.MET[BL].pt),
ak.to_numpy(ak.max(mjf[BL], axis=1)),
ak.to_numpy(pad_and_flatten(delta_eta[BL])),
ak.to_numpy(pad_and_flatten(leading_lepton[BL].pt)),
ak.to_numpy(pad_and_flatten(leading_lepton[BL].eta)),
ak.to_numpy(pad_and_flatten(trailing_lepton[BL].pt)),
ak.to_numpy(pad_and_flatten(trailing_lepton[BL].eta)),
ak.to_numpy(pad_and_flatten(dilepton_mass[BL])),
ak.to_numpy(pad_and_flatten(dilepton_pt[BL])),
ak.to_numpy(pad_and_flatten(j_fwd[BL].pt)),
ak.to_numpy(pad_and_flatten(j_fwd[BL].p)),
ak.to_numpy(pad_and_flatten(j_fwd[BL].eta)),
ak.to_numpy(pad_and_flatten(jet[:, 0:1][BL].pt)),
ak.to_numpy(pad_and_flatten(jet[:, 1:2][BL].pt)),
ak.to_numpy(pad_and_flatten(jet[:, 0:1][BL].eta)),
ak.to_numpy(pad_and_flatten(jet[:, 1:2][BL].eta)),
ak.to_numpy(pad_and_flatten(high_score_btag[:, 0:1][BL].pt)),
ak.to_numpy(pad_and_flatten(high_score_btag[:, 1:2][BL].pt)),
ak.to_numpy(pad_and_flatten(high_score_btag[:, 0:1][BL].eta)),
ak.to_numpy(pad_and_flatten(high_score_btag[:, 1:2][BL].eta)),
ak.to_numpy(min_bl_dR[BL]),
ak.to_numpy(min_mt_lep_met[BL]),
])
NN_inputs = np.moveaxis(NN_inputs, 0, 1)
model, scaler = load_onnx_model('v8')
try:
NN_inputs_scaled = scaler.transform(NN_inputs)
NN_pred = predict_onnx(model, NN_inputs_scaled)
best_score = np.argmax(NN_pred, axis=1)
except ValueError:
#print ("Empty NN_inputs")
NN_pred = np.array([])
best_score = np.array([])
NN_inputs_scaled = NN_inputs
#k.clear_session()
output['node'].fill(dataset=dataset, multiplicity=best_score, weight=weight_BL)
output['node0_score_incl'].fill(dataset=dataset, score=NN_pred[:,0] if np.shape(NN_pred)[0]>0 else np.array([]), weight=weight_BL)
output['node0_score'].fill(dataset=dataset, score=NN_pred[best_score==0][:,0] if np.shape(NN_pred)[0]>0 else np.array([]), weight=weight_BL[best_score==0])
output['node1_score'].fill(dataset=dataset, score=NN_pred[best_score==1][:,1] if np.shape(NN_pred)[0]>0 else np.array([]), weight=weight_BL[best_score==1])
output['node2_score'].fill(dataset=dataset, score=NN_pred[best_score==2][:,2] if np.shape(NN_pred)[0]>0 else np.array([]), weight=weight_BL[best_score==2])
output['node3_score'].fill(dataset=dataset, score=NN_pred[best_score==3][:,3] if np.shape(NN_pred)[0]>0 else np.array([]), weight=weight_BL[best_score==3])
output['node4_score'].fill(dataset=dataset, score=NN_pred[best_score==4][:,4] if np.shape(NN_pred)[0]>0 else np.array([]), weight=weight_BL[best_score==4])
SR_sel_pp = ((best_score==0) & ak.flatten((leading_lepton[BL].pdgId<0)))
SR_sel_mm = ((best_score==0) & ak.flatten((leading_lepton[BL].pdgId>0)))
leading_lepton_BL = leading_lepton[BL]
output['lead_lep_SR_pp'].fill(
dataset = dataset,
pt = ak.to_numpy(ak.flatten(leading_lepton_BL[SR_sel_pp].pt)),
weight = weight_BL[SR_sel_pp]
)
output['lead_lep_SR_mm'].fill(
dataset = dataset,
pt = ak.to_numpy(ak.flatten(leading_lepton_BL[SR_sel_mm].pt)),
weight = weight_BL[SR_sel_mm]
)
del model
del scaler
del NN_inputs, NN_inputs_scaled, NN_pred
# first, make a few super inclusive plots
output['PV_npvs'].fill(dataset=dataset, multiplicity=ev.PV[BL].npvs, weight=weight_BL)
output['PV_npvsGood'].fill(dataset=dataset, multiplicity=ev.PV[BL].npvsGood, weight=weight_BL)
output['N_jet'].fill(dataset=dataset, multiplicity=ak.num(jet)[BL], weight=weight_BL)
output['N_b'].fill(dataset=dataset, multiplicity=ak.num(btag)[BL], weight=weight_BL)
output['N_central'].fill(dataset=dataset, multiplicity=ak.num(central)[BL], weight=weight_BL)
output['N_ele'].fill(dataset=dataset, multiplicity=ak.num(electron)[BL], weight=weight_BL)
output['N_mu'].fill(dataset=dataset, multiplicity=ak.num(electron)[BL], weight=weight_BL)
output['N_fwd'].fill(dataset=dataset, multiplicity=ak.num(fwd)[BL], weight=weight_BL)
output['ST'].fill(dataset=dataset, pt=st[BL], weight=weight_BL)
output['HT'].fill(dataset=dataset, pt=ht[BL], weight=weight_BL)
if not re.search(re.compile('MuonEG|DoubleMuon|DoubleEG|EGamma'), dataset):
output['nLepFromTop'].fill(dataset=dataset, multiplicity=ev[BL].nLepFromTop, weight=weight_BL)
output['nLepFromTau'].fill(dataset=dataset, multiplicity=ev.nLepFromTau[BL], weight=weight_BL)
output['nLepFromZ'].fill(dataset=dataset, multiplicity=ev.nLepFromZ[BL], weight=weight_BL)
output['nLepFromW'].fill(dataset=dataset, multiplicity=ev.nLepFromW[BL], weight=weight_BL)
output['nGenTau'].fill(dataset=dataset, multiplicity=ev.nGenTau[BL], weight=weight_BL)
output['nGenL'].fill(dataset=dataset, multiplicity=ak.num(ev.GenL[BL], axis=1), weight=weight_BL)
output['chargeFlip_vs_nonprompt'].fill(dataset=dataset, n1=n_chargeflip[BL], n2=n_nonprompt[BL], n_ele=ak.num(electron)[BL], weight=weight_BL)
output['MET'].fill(
dataset = dataset,
pt = ev.MET[BL].pt,
phi = ev.MET[BL].phi,
weight = weight_BL
)
if not re.search(re.compile('MuonEG|DoubleMuon|DoubleEG|EGamma'), dataset):
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_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_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_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_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_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_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_BL
)
output['fwd_jet'].fill(
dataset = dataset,
pt = ak.flatten(j_fwd[BL].pt),
eta = ak.flatten(j_fwd[BL].eta),
phi = ak.flatten(j_fwd[BL].phi),
weight = weight_BL
)
output['high_p_fwd_p'].fill(dataset=dataset, p = ak.flatten(j_fwd[BL].p), weight = weight_BL)
return output
def process(self, events):
output = self.accumulator.identity()
dataset = events.metadata['dataset']
output['sumw'][dataset] += ak.sum(events.genWeight)
##############
# Trigger level
triggers = [
"HLT_Mu12_TrkIsoVVL_Ele23_CaloIdL_TrackIdL_IsoVL_DZ",
"HLT_Mu23_TrkIsoVVL_Ele12_CaloIdL_TrackIdL_IsoVL_DZ",
]
trig_arrs = [events.HLT[_trig.strip("HLT_")] for _trig in triggers]
req_trig = np.zeros(len(events), dtype='bool')
for t in trig_arrs:
req_trig = req_trig | t
############
# Event level
## Muon cuts
# muon twiki: https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideMuonIdRun2
events.Muon = events.Muon[(events.Muon.pt > 30) & (abs(events.Muon.eta < 2.4))] # & (events.Muon.tightId > .5)
events.Muon = ak.pad_none(events.Muon, 1, axis=1)
req_muon =(ak.count(events.Muon.pt, axis=1) == 1)
## Electron cuts
# electron twiki: https://twiki.cern.ch/twiki/bin/viewauth/CMS/CutBasedElectronIdentificationRun2
events.Electron = events.Electron[(events.Electron.pt > 30) & (abs(events.Electron.eta) < 2.4)]
events.Electron = ak.pad_none(events.Electron, 1, axis=1)
req_ele = (ak.count(events.Electron.pt, axis=1) == 1)
## Jet cuts
events.Jet = events.Jet[(events.Jet.pt > 25) & (abs(events.Jet.eta) <= 2.5)]
req_jets = (ak.count(events.Jet.pt, axis=1) >= 2)
req_opposite_charge = events.Electron[:, 0].charge * events.Muon[:, 0].charge == -1
event_level = req_trig & req_muon & req_ele & req_opposite_charge & req_jets
# Selected
selev = events[event_level]
#########
# Per electron
el_eta = (abs(selev.Electron.eta) <= 2.4)
el_pt = selev.Electron.pt > 30
el_level = el_eta & el_pt
# Per muon
mu_eta = (abs(selev.Muon.eta) <= 2.4)
mu_pt = selev.Muon.pt > 30
mu_level = mu_eta & mu_pt
# Per jet
jet_eta = (abs(selev.Jet.eta) <= 2.4)
jet_pt = selev.Jet.pt > 25
jet_pu = ( ((selev.Jet.puId > 6) & (selev.Jet.pt < 50)) | (selev.Jet.pt > 50) )
jet_id = selev.Jet.jetId >= 2
#jet_id = selev.Jet.isTight() == 1 & selev.Jet.isTightLeptonVeto() == 0
jet_level = jet_pu & jet_eta & jet_pt & jet_id
# b-tag twiki : https://twiki.cern.ch/twiki/bin/viewauth/CMS/BtagRecommendation102X
bjet_disc_t = selev.Jet.btagDeepB > 0.7264 # L=0.0494, M=0.2770, T=0.7264
bjet_disc_m = selev.Jet.btagDeepB > 0.2770 # L=0.0494, M=0.2770, T=0.7264
bjet_disc_l = selev.Jet.btagDeepB > 0.0494 # L=0.0494, M=0.2770, T=0.7264
bjet_level_t = jet_level & bjet_disc_t
bjet_level_m = jet_level & bjet_disc_m
bjet_level_l = jet_level & bjet_disc_l
sel = selev.Electron[el_level]
smu = selev.Muon[mu_level]
sjets = selev.Jet[jet_level]
sbjets_t = selev.Jet[bjet_level_t]
sbjets_m = selev.Jet[bjet_level_m]
sbjets_l = selev.Jet[bjet_level_l]
# output['pt'].fill(dataset=dataset, pt=selev.Jet.pt.flatten())
# Fill histograms dynamically
for histname, h in output.items():
if (histname not in self.jet_hists) and (histname not in self.deepcsv_hists): continue
# Get valid fields perhistogram to fill
fields = {k: ak.flatten(sjets[k], axis=None) for k in h.fields if k in dir(sjets)}
h.fill(dataset=dataset, **fields)
def flatten(ar): # flatten awkward into a 1d array to hist
return ak.flatten(ar, axis=None)
def num(ar):
return ak.num(ak.fill_none(ar[~ak.is_none(ar)], 0), axis=0)
output['njet'].fill(dataset=dataset, njet=flatten(ak.num(sjets)))
output['nbjet_t'].fill(dataset=dataset, nbjet_t=flatten(ak.num(sbjets_t)))
output['nbjet_m'].fill(dataset=dataset, nbjet_m=flatten(ak.num(sbjets_m)))
output['nbjet_l'].fill(dataset=dataset, nbjet_l=flatten(ak.num(sbjets_l)))
output['nel'].fill(dataset=dataset, nel=flatten(ak.num(sel)))
output['nmu'].fill(dataset=dataset, nmu=flatten(ak.num(smu)))
output['lelpt'].fill(dataset=dataset, lelpt=flatten(selev.Electron[:, 0].pt))
output['lmupt'].fill(dataset=dataset, lmupt=flatten(selev.Muon[:, 0].pt))
output['ljpt'].fill(dataset=dataset, ljpt=flatten(selev.Jet[:, 0].pt))
output['sljpt'].fill(dataset=dataset, sljpt=flatten(selev.Jet[:, 1].pt))
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 build_weight_like(weight, selection, like):
return ak.flatten(weight[selection] * ak.ones_like(like[selection]))
protons_ = select_protons(events_sel_, "ProtCandRnd")
counts_protons_ = len(protons_)
if not counts_label_protons_ in selections:
selections = np.concatenate(
(selections, np.array([counts_label_protons_])))
counts = np.concatenate((counts, np.array([counts_protons_])))
else:
counts[selections == counts_label_protons_] += counts_protons_
print(selections)
print(counts)
for col_ in columns:
protons_list[col_] = np.array(
ak.flatten(protons_[protons_keys[col_]]))
arr_size_ = len(protons_list["Xi"])
print("Flattened array size: {}".format(arr_size_))
dset_entries += arr_size_
if dset_entries > dset_chunk_size:
resize_factor_ = (dset_entries // dset_chunk_size)
chunk_resize_ = resize_factor_ * dset_chunk_size
print("Resizing output dataset by {} entries.".format(
chunk_resize_))
dset.resize((dset.shape[0] + chunk_resize_), axis=0)
print("Dataset shape: {}".format(dset.shape))
protons_ = None
if not random_protons_: protons_ = select_protons( events_sel_, "ProtCand" )
else: protons_ = select_protons( events_sel_, "ProtCandRnd" )
counts_protons_ = len( protons_ )
if not counts_label_protons_ in selections:
selections = np.concatenate( ( selections, np.array( [ counts_label_protons_ ] ) ) )
counts = np.concatenate( ( counts, np.array( [counts_protons_] ) ) )
else:
counts[ selections == counts_label_protons_ ] += counts_protons_
print ( selections )
print ( counts )
for col_ in columns:
protons_list[ col_ ] = np.array( ak.flatten( protons_[ protons_keys[ col_ ] ] ) )
arr_size_ = len( protons_list[ "Xi" ] )
print ( "Flattened array size: {}".format( arr_size_ ) )
dset_entries += arr_size_
if dset_entries > dset_chunk_size:
resize_factor_ = ( dset_entries // dset_chunk_size )
chunk_resize_ = resize_factor_ * dset_chunk_size
print ( "Resizing output dataset by {} entries.".format( chunk_resize_ ) )
dset.resize( ( dset.shape[0] + chunk_resize_ ), axis=0 )
print ( "Dataset shape: {}".format( dset.shape ) )
dset_slice += resize_factor_
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 flatten(ar): # flatten awkward into a 1d array to hist
return ak.flatten(ar, axis=None)
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
