def test_jagged_ufunc_table(self):
a = JaggedArray([0, 3, 3, 5], [3, 3, 5, 10], [0.0, 1.1, 2.2, 3.3, 4.4, 5.5, 6.6, 7.7, 8.8, 9.9])
assert (awkward.Table(x=[100, 200, 300, 400], y=[1000, 2000, 3000, 4000]) + a).tolist() == [{"x": [100.0, 101.1, 102.2], "y": [1000.0, 1001.1, 1002.2]}, {"x": [], "y": []}, {"x": [303.3, 304.4], "y": [3003.3, 3004.4]}, {"x": [405.5, 406.6, 407.7, 408.8, 409.9], "y": [4005.5, 4006.6, 4007.7, 4008.8, 4009.9]}]
a = JaggedArray([0, 3, 3, 5], [3, 3, 5, 10], awkward.Table(x=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9], y=[0.0, 1.1, 2.2, 3.3, 4.4, 5.5, 6.6, 7.7, 8.8, 9.9]))
assert (a + 1000).tolist() == [[{"x": 1000, "y": 1000.0}, {"x": 1001, "y": 1001.1}, {"x": 1002, "y": 1002.2}], [], [{"x": 1003, "y": 1003.3}, {"x": 1004, "y": 1004.4}], [{"x": 1005, "y": 1005.5}, {"x": 1006, "y": 1006.6}, {"x": 1007, "y": 1007.7}, {"x": 1008, "y": 1008.8}, {"x": 1009, "y": 1009.9}]]
assert (a + numpy.array([100, 200, 300, 400])).tolist() == [[{"x": 100, "y": 100.0}, {"x": 101, "y": 101.1}, {"x": 102, "y": 102.2}], [], [{"x": 303, "y": 303.3}, {"x": 304, "y": 304.4}], [{"x": 405, "y": 405.5}, {"x": 406, "y": 406.6}, {"x": 407, "y": 407.7}, {"x": 408, "y": 408.8}, {"x": 409, "y": 409.9}]]
assert (a + awkward.Table(x=[100, 200, 300, 400], y=[1000, 2000, 3000, 4000])).tolist() == [[{"x": 100, "y": 1000.0}, {"x": 101, "y": 1001.1}, {"x": 102, "y": 1002.2}], [], [{"x": 303, "y": 3003.3}, {"x": 304, "y": 3004.4}], [{"x": 405, "y": 4005.5}, {"x": 406, "y": 4006.6}, {"x": 407, "y": 4007.7}, {"x": 408, "y": 4008.8}, {"x": 409, "y": 4009.9}]]
def run_query(input_filenames=None, tree_name=None, branches=None):
import awkward, uproot
a = (lambda event: (awkward.Table if hasattr(awkward, 'Table') else awkward['Table'])((event.MVA3lCERN_weight_ttH if hasattr(event, 'MVA3lCERN_weight_ttH') else event['MVA3lCERN_weight_ttH'])))
b = (lambda event: event[(((((((((((((((((((((((((((event.trilep_type if hasattr(event, 'trilep_type') else event['trilep_type']) > 0) & ((event.nTaus_OR_Pt25 if hasattr(event, 'nTaus_OR_Pt25') else event['nTaus_OR_Pt25']) == 0)) & (abs((event.total_charge if hasattr(event, 'total_charge') else event['total_charge'])) == 1)) & ((event.nJets_OR_T if hasattr(event, 'nJets_OR_T') else event['nJets_OR_T']) >= 2)) & ((event.nJets_OR_T_MV2c10_70 if hasattr(event, 'nJets_OR_T_MV2c10_70') else event['nJets_OR_T_MV2c10_70']) > 0)) & ((event.lep_Pt_1 if hasattr(event, 'lep_Pt_1') else event['lep_Pt_1']) > 15000.0)) & ((event.lep_Pt_2 if hasattr(event, 'lep_Pt_2') else event['lep_Pt_2']) > 15000.0)) & ((event.lep_isolationFixedCutLoose_0 if hasattr(event, 'lep_isolationFixedCutLoose_0') else event['lep_isolationFixedCutLoose_0']) > 0)) & (abs(((event.Mlll012 if hasattr(event, 'Mlll012') else event['Mlll012']) - 91200.0)) > 10000.0)) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) != (-(event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1']))) | ((event.Mll01 if hasattr(event, 'Mll01') else event['Mll01']) > 12000.0))) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) != (-(event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2']))) | ((event.Mll02 if hasattr(event, 'Mll02') else event['Mll02']) > 12000.0))) & (((((abs((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0'])) == 13) & ((event.lep_isMedium_0 if hasattr(event, 'lep_isMedium_0') else event['lep_isMedium_0']) > 0)) | (abs((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0'])) == 11)) & (((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) == 11) & (abs((event.lep_Eta_1 if hasattr(event, 'lep_Eta_1') else event['lep_Eta_1'])) < 2.0)) | ((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) == 13) & ((event.lep_isMedium_1 if hasattr(event, 'lep_isMedium_1') else event['lep_isMedium_1']) > 0)))) & (((abs((event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2'])) == 11) & (abs((event.lep_Eta_2 if hasattr(event, 'lep_Eta_2') else event['lep_Eta_2'])) < 2.0)) | ((abs((event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2'])) == 13) & ((event.lep_isMedium_2 if hasattr(event, 'lep_isMedium_2') else event['lep_isMedium_2']) > 0))))) & ((((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) * abs((event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2']))) != 169) & ((event.DRll12 if hasattr(event, 'DRll12') else event['DRll12']) > 0.5)) | ((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) * abs((event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2']))) == 169))) & ((event.Mll12 if hasattr(event, 'Mll12') else event['Mll12']) > 12000.0)) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) != (-(event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1']))) | (abs(((event.Mll01 if hasattr(event, 'Mll01') else event['Mll01']) - 91200.0)) > 10000.0))) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) != (-(event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2']))) | (abs(((event.Mll02 if hasattr(event, 'Mll02') else event['Mll02']) - 91200.0)) > 10000.0))) & ((event.MVA3lCERN_weight_ttH if hasattr(event, 'MVA3lCERN_weight_ttH') else event['MVA3lCERN_weight_ttH']) > (-1))) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) != (-(event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1']))) | (abs(((event.Mll01 if hasattr(event, 'Mll01') else event['Mll01']) - 91200.0)) > 10000.0))) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) != (-(event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2']))) | (abs(((event.Mll02 if hasattr(event, 'Mll02') else event['Mll02']) - 91200.0)) > 10000.0))) & ((event.MVA3lCERN_weight_ttH if hasattr(event, 'MVA3lCERN_weight_ttH') else event['MVA3lCERN_weight_ttH']) > 0.3)) & ((event.MVA3lCERN_weight_ttW if hasattr(event, 'MVA3lCERN_weight_ttW') else event['MVA3lCERN_weight_ttW']) < 0.75)) & ((event.MVA3lCERN_weight_ttZ if hasattr(event, 'MVA3lCERN_weight_ttZ') else event['MVA3lCERN_weight_ttZ']) < 0.75)) & ((event.MVA3lCERN_weight_VV if hasattr(event, 'MVA3lCERN_weight_VV') else event['MVA3lCERN_weight_VV']) < 0.75)) & ((event.MVA3lCERN_weight_ttbar if hasattr(event, 'MVA3lCERN_weight_ttbar') else event['MVA3lCERN_weight_ttbar']) < 0.3)) & ((((((((event.dilep_type if hasattr(event, 'dilep_type') else event['dilep_type']) > 0) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) * (event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) > 0)) & ((event.lep_isQMisID_1 if hasattr(event, 'lep_isQMisID_1') else event['lep_isQMisID_1']) == 0)) & ((event.lep_isQMisID_0 if hasattr(event, 'lep_isQMisID_0') else event['lep_isQMisID_0']) == 0)) | ((((event.trilep_type if hasattr(event, 'trilep_type') else event['trilep_type']) > 0) & ((event.lep_isQMisID_2 if hasattr(event, 'lep_isQMisID_2') else event['lep_isQMisID_2']) == 0)) & ((event.lep_isQMisID_1 if hasattr(event, 'lep_isQMisID_1') else event['lep_isQMisID_1']) == 0))) | (((event.quadlep_type if hasattr(event, 'quadlep_type') else event['quadlep_type']) > 0) & ((event.FSF_4L_tot if hasattr(event, 'FSF_4L_tot') else event['FSF_4L_tot']) == 1))) & (((((((((event.dilep_type if hasattr(event, 'dilep_type') else event['dilep_type']) > 0) & (((((abs((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0'])) == 13) & ((event.lep_isMedium_0 if hasattr(event, 'lep_isMedium_0') else event['lep_isMedium_0']) > 0)) & ((event.lep_isolationFixedCutLoose_0 if hasattr(event, 'lep_isolationFixedCutLoose_0') else event['lep_isolationFixedCutLoose_0']) > 0)) & ((event.lep_promptLeptonVeto_TagWeight_0 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_0') else event['lep_promptLeptonVeto_TagWeight_0']) < (-0.5))) | ((((((abs((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0'])) == 11) & ((event.lep_isolationFixedCutLoose_0 if hasattr(event, 'lep_isolationFixedCutLoose_0') else event['lep_isolationFixedCutLoose_0']) > 0)) & ((event.lep_isTightLH_0 if hasattr(event, 'lep_isTightLH_0') else event['lep_isTightLH_0']) > 0)) & ((event.lep_chargeIDBDTTight_0 if hasattr(event, 'lep_chargeIDBDTTight_0') else event['lep_chargeIDBDTTight_0']) > 0.7)) & ((event.lep_ambiguityType_0 if hasattr(event, 'lep_ambiguityType_0') else event['lep_ambiguityType_0']) == 0)) & ((event.lep_promptLeptonVeto_TagWeight_0 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_0') else event['lep_promptLeptonVeto_TagWeight_0']) < (-0.7))))) & (((((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) == 13) & ((event.lep_isMedium_1 if hasattr(event, 'lep_isMedium_1') else event['lep_isMedium_1']) > 0)) & ((event.lep_isolationFixedCutLoose_1 if hasattr(event, 'lep_isolationFixedCutLoose_1') else event['lep_isolationFixedCutLoose_1']) > 0)) & ((event.lep_promptLeptonVeto_TagWeight_1 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_1') else event['lep_promptLeptonVeto_TagWeight_1']) < (-0.5))) | ((((((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) == 11) & ((event.lep_isolationFixedCutLoose_1 if hasattr(event, 'lep_isolationFixedCutLoose_1') else event['lep_isolationFixedCutLoose_1']) > 0)) & ((event.lep_isTightLH_1 if hasattr(event, 'lep_isTightLH_1') else event['lep_isTightLH_1']) > 0)) & ((event.lep_chargeIDBDTTight_1 if hasattr(event, 'lep_chargeIDBDTTight_1') else event['lep_chargeIDBDTTight_1']) > 0.7)) & ((event.lep_ambiguityType_1 if hasattr(event, 'lep_ambiguityType_1') else event['lep_ambiguityType_1']) == 0)) & ((event.lep_promptLeptonVeto_TagWeight_1 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_1') else event['lep_promptLeptonVeto_TagWeight_1']) < (-0.7))))) | ((((event.trilep_type if hasattr(event, 'trilep_type') else event['trilep_type']) > 0) & ((event.nTaus_OR_Pt25 if hasattr(event, 'nTaus_OR_Pt25') else event['nTaus_OR_Pt25']) == 0)) & (((((abs((event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2'])) == 13) & ((event.lep_isolationFixedCutLoose_2 if hasattr(event, 'lep_isolationFixedCutLoose_2') else event['lep_isolationFixedCutLoose_2']) > 0)) & ((event.lep_promptLeptonVeto_TagWeight_2 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_2') else event['lep_promptLeptonVeto_TagWeight_2']) < (-0.5))) | ((((((abs((event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2'])) == 11) & ((event.lep_isolationFixedCutLoose_2 if hasattr(event, 'lep_isolationFixedCutLoose_2') else event['lep_isolationFixedCutLoose_2']) > 0)) & ((event.lep_isTightLH_2 if hasattr(event, 'lep_isTightLH_2') else event['lep_isTightLH_2']) > 0)) & ((event.lep_chargeIDBDTTight_2 if hasattr(event, 'lep_chargeIDBDTTight_2') else event['lep_chargeIDBDTTight_2']) > 0.7)) & ((event.lep_promptLeptonVeto_TagWeight_2 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_2') else event['lep_promptLeptonVeto_TagWeight_2']) < (-0.7))) & ((event.lep_ambiguityType_2 if hasattr(event, 'lep_ambiguityType_2') else event['lep_ambiguityType_2']) == 0))) & ((((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) == 13) & ((event.lep_isolationFixedCutLoose_1 if hasattr(event, 'lep_isolationFixedCutLoose_1') else event['lep_isolationFixedCutLoose_1']) > 0)) & ((event.lep_promptLeptonVeto_TagWeight_1 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_1') else event['lep_promptLeptonVeto_TagWeight_1']) < (-0.5))) | ((((((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) == 11) & ((event.lep_isolationFixedCutLoose_1 if hasattr(event, 'lep_isolationFixedCutLoose_1') else event['lep_isolationFixedCutLoose_1']) > 0)) & ((event.lep_isTightLH_1 if hasattr(event, 'lep_isTightLH_1') else event['lep_isTightLH_1']) > 0)) & ((event.lep_chargeIDBDTTight_1 if hasattr(event, 'lep_chargeIDBDTTight_1') else event['lep_chargeIDBDTTight_1']) > 0.7)) & ((event.lep_promptLeptonVeto_TagWeight_1 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_1') else event['lep_promptLeptonVeto_TagWeight_1']) < (-0.7))) & ((event.lep_ambiguityType_1 if hasattr(event, 'lep_ambiguityType_1') else event['lep_ambiguityType_1']) == 0)))))) | (((event.dilep_type if hasattr(event, 'dilep_type') else event['dilep_type']) > 0) & ((event.nTaus_OR_Pt25 if hasattr(event, 'nTaus_OR_Pt25') else event['nTaus_OR_Pt25']) > 1))) | ((((event.dilep_type if hasattr(event, 'dilep_type') else event['dilep_type']) > 0) | ((event.trilep_type if hasattr(event, 'trilep_type') else event['trilep_type']) > 0)) == 0)) | ((event.quadlep_type if hasattr(event, 'quadlep_type') else event['quadlep_type']) > 0)) | (((event.trilep_type if hasattr(event, 'trilep_type') else event['trilep_type']) > 0) & ((event.nTaus_OR_Pt25 if hasattr(event, 'nTaus_OR_Pt25') else event['nTaus_OR_Pt25']) > 0))))) | ((event.is1L2Tau if hasattr(event, 'is1L2Tau') else event['is1L2Tau']) > 0))])
out = awkward.Table()
out['0'] =[]
for i in uproot.iterate(input_filenames,tree_name,branches=branches,namedecode="utf-8",entrysteps=50000, reportentries=False):
out = awkward.concatenate([out, (a)((b)(awkward.Table(i)))])
# for i in uproot.iterate(input_filenames,tree_name,branches=branches,namedecode="utf-8",entrysteps=10000, reportentries=True):
# print("Entry range: ", i[0], i[1])
# out = awkward.concatenate([out, (a)((b)(awkward.Table(i[2])))])
return out
def __getitem__(self, key):
if key in self.cache_:
return self.table_[key]
elif key in self.keys_:
ret = self.array(key)
self.table_[key] = ret
self.cache_.add(key)
return self.table_[key]
else:
branch = key + '_'
subset = [k for k in self.keys_ if k.startswith(branch)]
info = {i.replace(branch, ''): self.array(i) for i in subset}
counter = 'n' + key
counts = 0
if counter in self.keys_:
counts = self.array(counter)
for name, branch in info.items():
if not (branch.counts == counts).all():
raise ValueError(
f'Key {name} does not have the right shape')
info = {i: j.content for i, j in info.items()}
#check that everything is there to make a p4
if all(i in info for i in ['pt', 'eta', 'phi', 'mass'
]): # FIXME! wrong logic
ret = JaggedCandidateArray.candidatesfromcounts(counts, **info)
else:
ret = awk.Table(**info)
if all(i in info for i in ['pt', 'phi']):
ret['p4'] = uproot_methods.TLorentzVectorArray.from_ptetaphi(
ret['pt'], 0, ret['phi'], 0)
self.table_[key] = ret
self.cache_.add(key)
return self.table_[key]
def table(self):
out = awkward.Table()
out._view = self._view
out._base = self._base
out._rowname = self._rowname
out._contents = self._contents
return out
def arrow_table(self, chunk_size, event_limit=sys.maxint):
def group(iterator, n):
"""
Batch together chunks of events into a single yield
:param iterator: Iterator from which events are drown
:param n: Number of events to include in each yield
:return: Yields a list of n or fewer events
"""
done = False
while not done:
results = []
try:
for i in range(n):
results.append(iterator.next())
yield results
except StopIteration:
done = True
yield results
for events in group(self.event_iterator.iterate(event_limit),
chunk_size):
object_array = awkward.fromiter(events)
attr_dict = {}
for attr_name in self.event_iterator.attr_name_list:
branch_name = attr_name.split('.')[0].strip(' ')
a_name = attr_name.split('.')[1]
attr_dict[branch_name + '_' + a_name.strip('()')] = \
object_array[branch_name][a_name]
object_table = awkward.Table(**attr_dict)
yield awkward.toarrow(object_table)
def from_p3(cls, p3, t):
out = cls.__new__(cls)
out._initObjectArray(awkward.Table())
out["fX"] = p3.x
out["fY"] = p3.y
out["fZ"] = p3.z
out["fE"] = t
return out
def jaggedFromColumnGroup(cgroup):
if isinstance(cgroup,PhysicalColumnGroup):
return JaggedCandidateArray.candidatesfromcounts(counts = cgroup.counts(),
p4 = cgroup.p4Column(),
**cgroup.otherColumns())
else:
return awkward.JaggedArray.fromcounts(cgroup.counts(),
awkward.Table(cgroup.columns()))
def candidatesfromoffsets(cls,offsets,p4,**kwargs):
items = {}
if isinstance(p4,uproot_methods.TLorentzVectorArray):
items['p4'] = p4
else:
items['p4'] = uproot_methods.TLorentzVectorArray(p4[:,0],p4[:,1],
p4[:,2],p4[:,3])
thep4 = items['p4']
items['__fast_pt'] = fast_pt(thep4)
items['__fast_eta'] = fast_eta(thep4)
items['__fast_phi'] = fast_phi(thep4)
items['__fast_mass'] = fast_mass(thep4)
items.update(kwargs)
return cls.fromoffsets(offsets,awkward.Table(items))
def __getitem__(self, key):
if key in self.cache_:
return self.table_[key]
elif key in self.keys_:
ret = self.array(key)
self.table_[key] = ret
self.cache_.add(key)
return self.table_[key]
else:
branch = key + '_'
subset = [k for k in self.keys_ if k.startswith(branch)]
info = {i.replace(branch, ''): self.array(i) for i in subset}
counter = 'n' + key
counts = None
if counter in self.keys_:
counts = self.array(counter)
elif all(isinstance(i, awk.JaggedArray) for i in
info.values()): # In case counter is missing by mistake
print(
f'You probably forgot to ask to load {counter} as a branch. Inferring it...'
)
counts = info[list(info.keys())[0]].counts
if counts is not None:
for name, branch in info.items():
if not (branch.counts == counts).all():
raise ValueError(
f'Key {name} does not have the right shape')
#check that everything is there to make a p4
if counts is not None and all(
i in info for i in ['pt', 'eta', 'phi', 'mass']):
# flatted to use candidatesfromcounts, better option available?
info = {i: j.content for i, j in info.items()}
ret = JaggedCandidateArray.candidatesfromcounts(counts, **info)
elif counts is not None: # Not enough to make a JaggedCandidateArray, but a jagged table
ret = awk.JaggedArray.zip(**info)
else: # flat object
ret = awk.Table(**info)
# check if p4 can be made, with MET fix for missing eta and mass
if all(i in info for i in ['pt', 'phi']):
ret['p4'] = uproot_methods.TLorentzVectorArray.from_ptetaphi(
info['pt'], info.get('eta', 0), info['phi'],
info.get('mass', 0))
self.table_[key] = ret
self.cache_.add(key)
return self.table_[key]
def expand_chunk(chunk):
chunk = dict(chunk.iloc[0])
dataset = chunk.pop('_id_dataset')
chunkindex = chunk.pop('_id_chunkindex')
columnnames = set(k[8:-5] for k in chunk.keys()
if k.startswith('columns_') and k.endswith('_data'))
columns = {}
for name in columnnames:
columns[name] = unpack({
'size': chunk['columns_%s_size' % name],
'dtype': chunk['columns_%s_dtype' % name],
'data': chunk['columns_%s_data' % name],
})
table = awkward.Table(columns)
table = awkward.toarrow(table).to_pandas()
table['dataset'] = dataset
return table
def __init__(self, *infiles, branches=[]):
if all(isinstance(i, dict) for i in infiles):
self.tts = infiles
self.keys_ = set(self.tts[0].keys())
self.dict_like_ = True
elif all(not isinstance(i, dict) for i in infiles):
self.ufs = map(uproot.open, infiles)
self.tts = [i['Events'] for i in self.ufs]
self.keys_ = set([i.decode() for i in self.tts[0].keys()])
self.dict_like_ = False
else:
raise RuntimeError('Cannot mix files and dicts!')
if branches:
self.keys_ = set(i for i in self.keys_ if any(
fnmatch(i, branch) for branch in branches))
self.cache_ = set()
self.used_branches_ = set()
self.table_ = awk.Table()
def recurse(layout):
if isinstance(layout, awkward1.partition.PartitionedArray):
return awkward0.ChunkedArray(
[recurse(x) for x in layout.partitions])
elif isinstance(layout, awkward1.layout.NumpyArray):
return numpy.asarray(layout)
elif isinstance(layout, awkward1.layout.EmptyArray):
return numpy.array([])
elif isinstance(layout, awkward1.layout.RegularArray):
# content, size
if keeplayout:
raise ValueError(
"awkward0 has no equivalent of RegularArray; "
"try keeplayout=False")
offsets = numpy.arange(0,
(len(layout) + 1)*layout.size,
layout.size)
return awkward0.JaggedArray.fromoffsets(
offsets, recurse(layout.content))
elif isinstance(layout, awkward1.layout.ListArray32):
# starts, stops, content
return awkward0.JaggedArray(
numpy.asarray(layout.starts),
numpy.asarray(layout.stops),
recurse(layout.content))
elif isinstance(layout, awkward1.layout.ListArrayU32):
# starts, stops, content
return awkward0.JaggedArray(
numpy.asarray(layout.starts),
numpy.asarray(layout.stops),
recurse(layout.content))
elif isinstance(layout, awkward1.layout.ListArray64):
# starts, stops, content
return awkward0.JaggedArray(
numpy.asarray(layout.starts),
numpy.asarray(layout.stops),
recurse(layout.content))
elif isinstance(layout, awkward1.layout.ListOffsetArray32):
# offsets, content
return awkward0.JaggedArray.fromoffsets(
numpy.asarray(layout.offsets),
recurse(layout.content))
elif isinstance(layout, awkward1.layout.ListOffsetArrayU32):
# offsets, content
return awkward0.JaggedArray.fromoffsets(
numpy.asarray(layout.offsets),
recurse(layout.content))
elif isinstance(layout, awkward1.layout.ListOffsetArray64):
# offsets, content
return awkward0.JaggedArray.fromoffsets(
numpy.asarray(layout.offsets),
recurse(layout.content))
elif isinstance(layout, awkward1.layout.Record):
# istuple, numfields, field(i)
out = []
for i in range(layout.numfields):
content = layout.field(i)
if isinstance(content, (awkward1.layout.Content,
awkward1.layout.Record)):
out.append(recurse(content))
else:
out.append(content)
if layout.istuple:
return tuple(out)
else:
return dict(zip(layout.keys(), out))
elif isinstance(layout, awkward1.layout.RecordArray):
# istuple, numfields, field(i)
if layout.numfields == 0 and len(layout) != 0:
raise ValueError(
"cannot convert zero-field, nonzero-length RecordArray "
"to awkward0.Table (limitation in awkward0)")
keys = layout.keys()
values = [recurse(x) for x in layout.contents]
pairs = collections.OrderedDict(zip(keys, values))
out = awkward0.Table(pairs)
if layout.istuple:
out._rowname = "tuple"
return out
elif isinstance(layout, awkward1.layout.UnionArray8_32):
# tags, index, numcontents, content(i)
return awkward0.UnionArray(numpy.asarray(layout.tags),
numpy.asarray(layout.index),
[recurse(x) for x in layout.contents])
elif isinstance(layout, awkward1.layout.UnionArray8_U32):
# tags, index, numcontents, content(i)
return awkward0.UnionArray(numpy.asarray(layout.tags),
numpy.asarray(layout.index),
[recurse(x) for x in layout.contents])
elif isinstance(layout, awkward1.layout.UnionArray8_64):
# tags, index, numcontents, content(i)
return awkward0.UnionArray(numpy.asarray(layout.tags),
numpy.asarray(layout.index),
[recurse(x) for x in layout.contents])
elif isinstance(layout, awkward1.layout.IndexedOptionArray32):
# index, content
index = numpy.asarray(layout.index)
toosmall = (index < -1)
if toosmall.any():
index = index.copy()
index[toosmall] = -1
return awkward0.IndexedMaskedArray(index, recurse(layout.content))
elif isinstance(layout, awkward1.layout.IndexedOptionArray64):
# index, content
index = numpy.asarray(layout.index)
toosmall = (index < -1)
if toosmall.any():
index = index.copy()
index[toosmall] = -1
return awkward0.IndexedMaskedArray(index, recurse(layout.content))
elif isinstance(layout, awkward1.layout.IndexedArray32):
# index, content
return awkward0.IndexedArray(numpy.asarray(layout.index),
recurse(layout.content))
elif isinstance(layout, awkward1.layout.IndexedArrayU32):
# index, content
return awkward0.IndexedArray(numpy.asarray(layout.index),
recurse(layout.content))
elif isinstance(layout, awkward1.layout.IndexedArray64):
# index, content
return awkward0.IndexedArray(numpy.asarray(layout.index),
recurse(layout.content))
elif isinstance(layout, awkward1.layout.ByteMaskedArray):
# mask, content, valid_when
return awkward0.MaskedArray(numpy.asarray(layout.mask),
recurse(layout.content),
maskedwhen=(not layout.valid_when))
elif isinstance(layout, awkward1.layout.BitMaskedArray):
# mask, content, valid_when, length, lsb_order
return awkward0.BitMaskedArray(numpy.asarray(layout.mask),
recurse(layout.content),
maskedwhen=(not layout.valid_when),
lsborder=layout.lsb_order)
elif isinstance(layout, awkward1.layout.UnmaskedArray):
# content
return recurse(layout.content) # no equivalent in awkward0
else:
raise AssertionError(
"missing converter for {0}".format(type(layout).__name__))
def Initialize(items):
argkeys = items.keys()
p4 = None
if 'p4' in argkeys:
p4 = items['p4']
if not isinstance(p4, uproot_methods.TLorentzVectorArray):
p4 = SaiyanArray.from_cartesian(p4[:, 0], p4[:, 1], p4[:, 2],
p4[:, 3])
elif 'pt' in argkeys and 'eta' in argkeys and 'phi' in argkeys and 'mass' in argkeys:
temp = SaiyanArray.from_ptetaphim(items['pt'], items['eta'],
items['phi'],
items['mass'])._to_cartesian()
p4 = SaiyanArray.from_cartesian(temp.x, temp.y, temp.z, temp.t)
del items['pt']
del items['eta']
del items['phi']
del items['mass']
elif 'pt' in argkeys and 'eta' in argkeys and 'phi' in argkeys and 'energy' in argkeys:
p4 = SaiyanArray.from_ptetaphi(items['pt'], items['eta'], items['phi'],
items['energy'])
del items['pt']
del items['eta']
del items['phi']
del items['energy']
elif 'px' in argkeys and 'py' in argkeys and 'pz' in argkeys and 'mass' in argkeys:
p4 = SaiyanArray.from_xyzm(items['px'], items['py'], items['pz'],
items['mass'])
del items['px']
del items['py']
del items['pz']
del items['mass']
elif 'pt' in argkeys and 'phi' in argkeys and 'pz' in argkeys and 'energy' in argkeys:
p4 = SaiyanArray.from_cylindrical(items['pt'], items['phi'],
items['pz'], items['energy'])
del items['pt']
del items['phi']
del items['pz']
del items['energy']
elif 'px' in argkeys and 'py' in argkeys and 'pz' in argkeys and 'energy' in argkeys:
p4 = SaiyanArray.from_cartesian(items['px'], items['py'], items['pz'],
items['energy'])
del items['px']
del items['py']
del items['pz']
del items['energy']
elif 'p' in argkeys and 'theta' in argkeys and 'phi' in argkeys and 'energy' in argkeys:
p4 = SaiyanArray.from_spherical(items['p'], items['theta'],
items['phi'], items['energy'])
del items['p']
del items['theta']
del items['phi']
del items['energy']
elif 'p3' in argkeys and 'energy' in argkeys:
p4 = SaiyanArray.from_p3(items['p3'], items['energy'])
del items['p3']
del items['energy']
try:
p4
except NameError:
out = awkward.Table()
else:
out = p4
for name, value in items.items():
out[name] = value
if isinstance(out, awkward.JaggedArray):
out = JaggedSaiyanArrayMethods.fromjagged(out)
else: # p4 is not None and not isinstance(out, awkward.JaggedArray):
try:
p4
except NameError:
out.__class__ = type("Event", (out.__class__, Dangerousness), {})
else:
out.__class__ = type("Object",
(out.__class__, SaiyanCommonMethods), {})
return out
def __init__(self, x):
self._initObjectArray(awkward.Table())
self["x"] = x
def open_file(filepath, njets=None):
"""
returns `X`, `y`, numpy arrays, in a format required by energyflow
filepath:
string, path to the .root file you wish to open
- `X` : a three-dimensional numpy array of the jets with shape
`(num_data,max_num_particles,5)`. 5: pt, eta, phi, E, tag
- `y` : binary array, 1 for real b-jets
"""
# open the file
sm_file = ur.open(filepath)
# get all data in tree format
sm_tree = sm_file['XhhMiniNtuple']
# get branches we care about
branches = sm_tree.arrays(branches=[
'resolvedJets_pt', 'resolvedJets_eta', 'resolvedJets_phi',
'resolvedJets_E', 'resolvedJets_HadronConeExclTruthLabelID',
'resolvedJets_is_DL1r_FixedCutBEff_77'
],
namedecode='utf-8')
# Meanings of the branches we took
"""
resolvedJets_pt:
transverse momentum of each jet
resolvedJets_eta:
pseudorapidity of each jet
resolvedJets_phi:
azimuth of each jet (angle around beam)
resolvedJets_E:
energy of each jet
resolvedJets_HadronConeExclTruthLabelID:
see Monte Carlo paper, classification number, e.g. 5=bjet, 15=tau
resolvedJets_is_DL1r_FixedCutBEff_77:
whether or not a jet has been b-tagged
"""
# convert to "pandas-dataframe-style" table
table = awkward.Table(branches)
# use pt, eta, phi, E to make LorentzVectors for each jet
lv = urm.TLorentzVectorArray.from_ptetaphie(table['resolvedJets_pt'],
table['resolvedJets_eta'],
table['resolvedJets_phi'],
table['resolvedJets_E'])
# add LVs to table
table['resolved_lv'] = lv
# add a "truth" category to the table,
# based on whether or not a jet is a b-jet or not
table['truth'] = (table['resolvedJets_HadronConeExclTruthLabelID'] == 5)
table['truth'] = table['truth'].astype(np.int32)
# and for convenience rename resolvedJets_is_DL1r_FixedCutBEff_77 to "tag"
table['tag'] = table['resolvedJets_is_DL1r_FixedCutBEff_77']
# a few easier names to use later
# number of jets overall
table['njets'] = awkward.AwkwardArray.count(table.resolved_lv.pt)
# number of b jets in each event
table['nbjets'] = awkward.AwkwardArray.count_nonzero(table['truth'])
# number of b tags in each event
table['nbtags'] = awkward.AwkwardArray.count_nonzero(table['tag'])
# sort table by tag
indices = table["tag"].argsort()
s_table = awkward.Table()
for label in ['resolved_lv', 'truth', 'tag']:
s_table[label] = table[label][indices]
max_njets = max(table.njets)
print("padding arrays")
pt_arr = np.array([
np.concatenate((pt, [0] * (max_njets - len(pt))))
for pt in s_table.resolved_lv.pt
])
eta_arr = np.array([
np.concatenate((eta, [0] * (max_njets - len(eta))))
for eta in s_table.resolved_lv.eta
])
phi_arr = np.array([
np.concatenate((phi, [0] * (max_njets - len(phi))))
for phi in s_table.resolved_lv.phi
])
E_arr = np.array([
np.concatenate((e, [0] * (max_njets - len(e))))
for e in s_table.resolved_lv.E
])
tag_arr = np.array([
np.concatenate((tag, [0] * (max_njets - len(tag))))
for tag in s_table.tag
])
truth_arr = np.array([
np.concatenate((tru, [0] * (max_njets - len(tru))))
for tru in s_table.truth
])
print('done padding')
if njets:
pt_arr = pt_arr[:, :njets]
eta_arr = eta_arr[:, :njets]
phi_arr = phi_arr[:, :njets]
E_arr = E_arr[:, :njets]
tag_arr = tag_arr[:, :njets]
truth_arr = truth_arr[:, :njets]
# ensure the first 3 jets are tagged correctly
events = np.ones((len(pt_arr)), dtype=bool)
print(np.count_nonzero(events), 'events total')
events[tag_arr[:, 0] == 0] = 0
events[tag_arr[:, 1] == 0] = 0
events[tag_arr[:, 2] == 0] = 0
events[truth_arr[:, 0] == 0] = 0
events[truth_arr[:, 1] == 0] = 0
events[truth_arr[:, 2] == 0] = 0
print(np.count_nonzero(events),
'events after ensuring first 3 are correctly tagged')
# and ensure no other jets are tagged
events[np.count_nonzero(tag_arr[:, 3:], axis=1) > 0] = 0
print(np.count_nonzero(events),
'events after ensuring we only have 3 tags')
# 1 = should've been tagged, wasn't. 0 = correctly tagged (/ not tagged)
untagged = np.logical_xor(truth_arr, tag_arr).astype(int)
n_untagged = np.count_nonzero(untagged, axis=1)
# ensure we only have at max one untagged jet
events[n_untagged > 1] = 0
print(np.count_nonzero(events),
'events after ensuring there is at most 1 untagged jet')
# create our data with only the filtered events
pt_arr = pt_arr[events]
eta_arr = eta_arr[events]
phi_arr = phi_arr[events]
E_arr = E_arr[events]
tag_arr = tag_arr[events]
truth_arr = truth_arr[events]
if njets:
X = np.zeros((len(pt_arr), njets, 5), dtype=float)
else:
X = np.zeros((len(pt_arr), max_njets, 5), dtype=float)
X[:, :, 0] = pt_arr
X[:, :, 1] = eta_arr
X[:, :, 2] = phi_arr
X[:, :, 3] = E_arr
X[:, :, 4] = tag_arr # should be all zeros at this point
# where does the untagged jet occur?
missed_jet_events, missed_jet_index = np.where(untagged[events] == 1)
# njets = no 4th jet, the rest means set that index's jet = 4th jet
missed_jet = np.zeros(len(pt_arr), dtype=int)
missed_jet += njets if njets else max_njets # don't pick a jet
missed_jet[missed_jet_events] = missed_jet_index # unless you should
if njets:
missed_jet_arr = np.zeros((len(pt_arr), njets + 1), dtype=int)
else:
missed_jet_arr = np.zeros((len(pt_arr), max_njets + 1), dtype=int)
for i, m in enumerate(missed_jet):
missed_jet_arr[i][m] = 1
y = missed_jet_arr
return X, y
def __init__(self, x, y, z, t):
self._initObjectArray(awkward.Table())
self["fX"] = x
self["fY"] = y
self["fZ"] = z
self["fE"] = t
def open_file(filepath, sort_by=None, pt_cut=None, eta_cut=None):
"""
open_file returns an awkward table, sorted by the parameter sort_by
filepath:
string, path to the .root file you wish to open
sort_by:
string, parameter with which to sort the table,
currently only "tag", "phi", "eta", "pt" are options
"""
# open the file
sm_file = ur.open(filepath)
# get all data in tree format
sm_tree = sm_file['XhhMiniNtuple']
# get branches we care about
branches = sm_tree.arrays(branches=[
'resolvedJets_pt', 'resolvedJets_eta', 'resolvedJets_phi',
'resolvedJets_E', 'resolvedJets_HadronConeExclTruthLabelID',
'resolvedJets_is_DL1r_FixedCutBEff_77', 'mcEventWeight'
],
namedecode='utf-8')
# Meanings of the branches we took
"""
resolvedJets_pt:
transverse momentum of each jet
resolvedJets_eta:
pseudorapidity of each jet
resolvedJets_phi:
azimuth of each jet (angle around beam)
resolvedJets_E:
energy of each jet
resolvedJets_HadronConeExclTruthLabelID:
see Monte Carlo paper, classification number, e.g. 5=bjet, 15=tau
resolvedJets_is_DL1r_FixedCutBEff_77:
whether or not a jet has been b-tagged
"""
# convert to "pandas-dataframe-style" table
table = awkward.Table(branches)
# get important data
pt = table['resolvedJets_pt']
eta = table['resolvedJets_eta']
phi = table['resolvedJets_phi']
E = table['resolvedJets_E']
truth = (table['resolvedJets_HadronConeExclTruthLabelID'] == 5).astype(
np.int32)
tag = table['resolvedJets_is_DL1r_FixedCutBEff_77']
print(len(pt), 'total events found')
# apply cuts if needed
if pt_cut != None:
if pt_cut < 0:
cut = pt <= abs(pt_cut)
else:
cut = pt > pt_cut
pt = pt[cut]
eta = eta[cut]
phi = phi[cut]
E = E[cut]
truth = truth[cut]
tag = tag[cut]
#print(len(pt), 'events left after pt cut')
if eta_cut != None:
if eta_cut < 0:
cut = abs(eta) >= abs(eta_cut)
else:
cut = abs(eta) < eta_cut
pt = pt[cut]
eta = eta[cut]
phi = phi[cut]
E = E[cut]
truth = truth[cut]
tag = tag[cut]
#print(len(pt), 'events left after eta cut')
# use pt, eta, phi, E to make LorentzVectors for each jet
lv = urm.TLorentzVectorArray.from_ptetaphie(pt, eta, phi, E)
# add LVs to table
table['resolved_lv'] = lv
# add a "truth" category to the table,
# based on whether or not a jet is a b-jet or not
table['truth'] = truth
# and for convenience rename resolvedJets_is_DL1r_FixedCutBEff_77 to "tag"
table['tag'] = tag
# sort table if needed
if sort_by is not None:
print("sorting data by", sort_by)
# indices of table, sorted by whatever variable is provided
if sort_by == "pt":
indices = table['resolved_lv'].pt.argsort()
elif sort_by == "eta":
indices = table['resolved_lv'].eta.argsort()
elif sort_by == "phi":
indices = table['resolved_lv'].phi.argsort()
elif sort_by == "tag":
indices = table["tag"].argsort()
else:
raise ValueError(f"sort_by={sort_by} is not yet supported")
# make new sorted table with attributes we need, properly sorted
s_table = awkward.Table()
for label in ['resolved_lv', 'truth', 'tag']:
s_table[label] = table[label][indices]
s_table['mcEventWeight'] = table['mcEventWeight']
else:
print("not sorting data")
s_table = table
# a few easier names to use later
# number of jets overall
s_table['njets'] = awkward.AwkwardArray.count(pt)
# number of b jets in each event
s_table['nbjets'] = awkward.AwkwardArray.count_nonzero(truth)
# number of b tags in each event
s_table['nbtags'] = awkward.AwkwardArray.count_nonzero(tag)
return s_table
def __init__(self, x, y):
self._initObjectArray(awkward.Table())
self["fX"] = x
self["fY"] = y
import pandas as pd
import backports.lzma
from tqdm.auto import tqdm
if __name__ == "__main__":
os.system("mkdir -p tables/")
os.system("mkdir -p jsons/")
f = uproot.open("doublemu_lzma.root")
t = f["Events"]
raw = t.arrays(["/Muon_(pt|eta|phi|mass|charge)/"],
outputtype=dict,
namedecode="ascii")
table = awkward.Table(raw)
print(table.nbytes / 1e6)
for i in range(20):
data = []
for fcomp, fdecomp, label in tqdm([
[
lambda x: backports.lzma.compress(x),
backports.lzma.decompress, "lzma"
],
[lambda x: blosc.compress(x), blosc.decompress,
"blosc"], # default should be blosc.SHUFFLE
[
lambda x: blosc.compress(x, shuffle=blosc.NOSHUFFLE),
awkward.JaggedArray.fromoffsets(offsets, arrays[b"Electron_phi"].content),
"mass":
awkward.JaggedArray.fromoffsets(offsets, arrays[b"Electron_mass"].content),
"cutBased":
awkward.JaggedArray.fromoffsets(offsets,
arrays[b"Electron_cutBased"].content),
"pdgId":
awkward.JaggedArray.fromoffsets(offsets,
arrays[b"Electron_pdgId"].content),
"pfRelIso03_all":
awkward.JaggedArray.fromoffsets(offsets,
arrays[b"Electron_pfRelIso03_all"].content)
}
electrons = physics_objects["Electron"]
t = awkward.Table(electrons)
awkward.toparquet("tmp.parquet", t)
pa_table = awkward.toarrow(t)
batches = pa_table.to_batches()
batch = batches[0]
sink = pa.BufferOutputStream()
writer = pa.RecordBatchStreamWriter(sink, batch.schema)
writer.write_batch(batch)
writer.close()
buf = sink.getvalue()
reader = pa.ipc.open_stream(buf)
batches2 = [b for b in reader]
def from_collections(wpartons_up=None,
wpartons_dw=None,
charged_leps=None,
neutral_leps=None,
bs=None,
bbars=None,
wbosons=None,
tops=None,
tbars=None):
#set_trace()
dilep_evts = ((charged_leps.counts == 2) &
(charged_leps.counts == neutral_leps.counts) &
(charged_leps.charge.sum() == 0))
dihad_evts = ((wpartons_up.counts == 2) &
(wpartons_up.counts == wpartons_dw.counts))
semilep_evts = ((charged_leps.counts == 1) & (neutral_leps.counts == 1) &
(wpartons_up.counts == 1) & (wpartons_dw.counts == 1))
valid_evts = (dilep_evts | dihad_evts | semilep_evts)
if not (valid_evts.sum() == tbars.size):
raise ValueError("Not all ttbar events are valid!")
dl_array = np.repeat(dilep_evts, valid_evts.astype(int) * 2)
dh_array = np.repeat(dihad_evts, valid_evts.astype(int) * 2)
sl_array = np.repeat(semilep_evts, valid_evts.astype(int) * 2)
if (bs[valid_evts].counts != 1).any():
raise ValueError("Number of b partons in valid events != 1")
if (bbars[valid_evts].counts != 1).any():
raise ValueError("Number of bbar partons in valid events != 1")
if (tops[valid_evts].counts != 1).any():
raise ValueError("Number of top partons in valid events != 1")
if (tbars[valid_evts].counts != 1).any():
raise ValueError("Number of tbar partons in valid events != 1")
# create leading/subleading objects from wpartons
# init vars
Lpt = np.ones(wpartons_up.counts.sum())
Leta = np.ones(wpartons_up.counts.sum())
Lphi = np.ones(wpartons_up.counts.sum())
Lmass = np.ones(wpartons_up.counts.sum())
Lcharge = np.zeros(wpartons_up.counts.sum())
LpdgId = np.zeros(wpartons_up.counts.sum())
Ldecaytype = np.ones(wpartons_up.counts.sum()) * 2
Spt = np.ones(wpartons_up.counts.sum())
Seta = np.ones(wpartons_up.counts.sum())
Sphi = np.ones(wpartons_up.counts.sum())
Smass = np.ones(wpartons_up.counts.sum())
Scharge = np.zeros(wpartons_up.counts.sum())
SpdgId = np.zeros(wpartons_up.counts.sum())
up_isLeading = (wpartons_up.pt > wpartons_dw.pt).flatten()
# Leading
Lpt = np.where(up_isLeading, wpartons_up.pt.flatten(),
wpartons_dw.pt.flatten())
Leta = np.where(up_isLeading, wpartons_up.eta.flatten(),
wpartons_dw.eta.flatten())
Lphi = np.where(up_isLeading, wpartons_up.phi.flatten(),
wpartons_dw.phi.flatten())
Lmass = np.where(up_isLeading, wpartons_up.mass.flatten(),
wpartons_dw.mass.flatten())
Lcharge = np.where(up_isLeading, wpartons_up.charge.flatten(),
wpartons_dw.charge.flatten())
LpdgId = np.where(up_isLeading, wpartons_up.pdgId.flatten(),
wpartons_dw.pdgId.flatten())
First = JaggedCandidateArray.candidatesfromcounts(
counts=wpartons_up.counts,
pt=Lpt,
eta=Leta,
phi=Lphi,
mass=Lmass,
charge=Lcharge,
pdgId=LpdgId,
decaytype=Ldecaytype,
)
# Subleading
Spt = np.where(~up_isLeading, wpartons_up.pt.flatten(),
wpartons_dw.pt.flatten())
Seta = np.where(~up_isLeading, wpartons_up.eta.flatten(),
wpartons_dw.eta.flatten())
Sphi = np.where(~up_isLeading, wpartons_up.phi.flatten(),
wpartons_dw.phi.flatten())
Smass = np.where(~up_isLeading, wpartons_up.mass.flatten(),
wpartons_dw.mass.flatten())
Scharge = np.where(~up_isLeading, wpartons_up.charge.flatten(),
wpartons_dw.charge.flatten())
SpdgId = np.where(~up_isLeading, wpartons_up.pdgId.flatten(),
wpartons_dw.pdgId.flatten())
Second = JaggedCandidateArray.candidatesfromcounts(
counts=wpartons_up.counts,
pt=Spt,
eta=Seta,
phi=Sphi,
mass=Smass,
charge=Scharge,
pdgId=SpdgId,
decaytype=Ldecaytype,
)
#set_trace()
# make objects from top/tbar decays
GenB = JaggedCandidateArray.candidatesfromcounts(
counts=valid_evts.astype(int),
pt=bs[valid_evts].pt.flatten(),
eta=bs[valid_evts].eta.flatten(),
phi=bs[valid_evts].phi.flatten(),
mass=bs[valid_evts].mass.flatten(),
charge=bs[valid_evts].charge.flatten(),
decaytype=wbosons[wbosons.charge == 1].decaytype.flatten(),
)
GenTop = JaggedCandidateArray.candidatesfromcounts(
counts=valid_evts.astype(int),
pt=tops[valid_evts].pt.flatten(),
eta=tops[valid_evts].eta.flatten(),
phi=tops[valid_evts].phi.flatten(),
mass=tops[valid_evts].mass.flatten(),
charge=tops[valid_evts].charge.flatten(),
decaytype=wbosons[wbosons.charge == 1].decaytype.flatten(),
)
GenBbar = JaggedCandidateArray.candidatesfromcounts(
counts=valid_evts.astype(int),
pt=bbars[valid_evts].pt.flatten(),
eta=bbars[valid_evts].eta.flatten(),
phi=bbars[valid_evts].phi.flatten(),
mass=bbars[valid_evts].mass.flatten(),
charge=bbars[valid_evts].charge.flatten(),
decaytype=wbosons[wbosons.charge == -1].decaytype.flatten(),
)
GenTbar = JaggedCandidateArray.candidatesfromcounts(
counts=valid_evts.astype(int),
pt=tbars[valid_evts].pt.flatten(),
eta=tbars[valid_evts].eta.flatten(),
phi=tbars[valid_evts].phi.flatten(),
mass=tbars[valid_evts].mass.flatten(),
charge=tbars[valid_evts].charge.flatten(),
decaytype=wbosons[wbosons.charge == -1].decaytype.flatten(),
)
ttbars = (GenTop.p4 + GenTbar.p4)
GenTTbar = JaggedCandidateArray.candidatesfromcounts(
counts=valid_evts.astype(int),
pt=ttbars.pt.flatten(),
eta=ttbars.eta.flatten(),
phi=ttbars.phi.flatten(),
mass=ttbars.mass.flatten(),
decaytype=wbosons[valid_evts].decaytype.sum(
), # 0 is for INVALID, 2 for DILEP, 3 for SEMILEP, 4 for HADRONIC
)
#set_trace()
## make tables of TTbar event objects
DILEP_evts = awkward.Table(
TTbar=awkward.JaggedArray.fromcounts(dilep_evts.astype(int),
GenTTbar[dilep_evts].flatten()),
Top=awkward.JaggedArray.fromcounts(dilep_evts.astype(int),
GenTop[dilep_evts].flatten()),
Tbar=awkward.JaggedArray.fromcounts(dilep_evts.astype(int),
GenTbar[dilep_evts].flatten()),
B=awkward.JaggedArray.fromcounts(dilep_evts.astype(int),
GenB[dilep_evts].flatten()),
Bbar=awkward.JaggedArray.fromcounts(dilep_evts.astype(int),
GenBbar[dilep_evts].flatten()),
Wplus=awkward.JaggedArray.fromcounts(
dilep_evts.astype(int),
wbosons[wbosons.charge == 1][dilep_evts].flatten()),
Wminus=awkward.JaggedArray.fromcounts(
dilep_evts.astype(int),
wbosons[wbosons.charge == -1][dilep_evts].flatten()),
First_plus=awkward.JaggedArray.fromcounts(
dilep_evts.astype(int),
charged_leps[dilep_evts][charged_leps[dilep_evts].charge > 0].
flatten()), # charged lepton always made leading
Second_plus=awkward.JaggedArray.fromcounts(
dilep_evts.astype(int),
neutral_leps[dilep_evts][charged_leps[dilep_evts].charge > 0].
flatten()), # neutral lepton always made subleading
First_minus=awkward.JaggedArray.fromcounts(
dilep_evts.astype(int),
charged_leps[dilep_evts][charged_leps[dilep_evts].charge < 0].
flatten()), # charged lepton always made leading
Second_minus=awkward.JaggedArray.fromcounts(
dilep_evts.astype(int),
neutral_leps[dilep_evts][charged_leps[dilep_evts].charge < 0].
flatten()), # neutral lepton always made subleading
Up_plus=awkward.JaggedArray.fromcounts(
dilep_evts.astype(int), neutral_leps[dilep_evts][
charged_leps[dilep_evts].charge > 0].flatten()),
Down_plus=awkward.JaggedArray.fromcounts(
dilep_evts.astype(int), charged_leps[dilep_evts][
charged_leps[dilep_evts].charge > 0].flatten()),
Up_minus=awkward.JaggedArray.fromcounts(
dilep_evts.astype(int), neutral_leps[dilep_evts][
charged_leps[dilep_evts].charge < 0].flatten()),
Down_minus=awkward.JaggedArray.fromcounts(
dilep_evts.astype(int), charged_leps[dilep_evts][
charged_leps[dilep_evts].charge < 0].flatten()),
)
HAD_evts = awkward.Table(
TTbar=awkward.JaggedArray.fromcounts(dihad_evts.astype(int),
GenTTbar[dihad_evts].flatten()),
Top=awkward.JaggedArray.fromcounts(dihad_evts.astype(int),
GenTop[dihad_evts].flatten()),
Tbar=awkward.JaggedArray.fromcounts(dihad_evts.astype(int),
GenTbar[dihad_evts].flatten()),
B=awkward.JaggedArray.fromcounts(dihad_evts.astype(int),
GenB[dihad_evts].flatten()),
Bbar=awkward.JaggedArray.fromcounts(dihad_evts.astype(int),
GenBbar[dihad_evts].flatten()),
Wplus=awkward.JaggedArray.fromcounts(
dihad_evts.astype(int),
wbosons[wbosons.charge == 1][dihad_evts].flatten()),
Wminus=awkward.JaggedArray.fromcounts(
dihad_evts.astype(int),
wbosons[wbosons.charge == -1][dihad_evts].flatten()),
First_plus=awkward.JaggedArray.fromcounts(
dihad_evts.astype(int),
First[First.charge > 0][dihad_evts].flatten()),
Second_plus=awkward.JaggedArray.fromcounts(
dihad_evts.astype(int),
Second[Second.charge > 0][dihad_evts].flatten()),
First_minus=awkward.JaggedArray.fromcounts(
dihad_evts.astype(int),
First[First.charge < 0][dihad_evts].flatten()),
Second_minus=awkward.JaggedArray.fromcounts(
dihad_evts.astype(int),
Second[Second.charge < 0][dihad_evts].flatten()),
Up_plus=awkward.JaggedArray.fromcounts(
dihad_evts.astype(int),
wpartons_up[wpartons_up.charge > 0][dihad_evts].flatten()),
Down_plus=awkward.JaggedArray.fromcounts(
dihad_evts.astype(int),
wpartons_dw[wpartons_dw.charge > 0][dihad_evts].flatten()),
Up_minus=awkward.JaggedArray.fromcounts(
dihad_evts.astype(int),
wpartons_up[wpartons_up.charge < 0][dihad_evts].flatten()),
Down_minus=awkward.JaggedArray.fromcounts(
dihad_evts.astype(int),
wpartons_dw[wpartons_dw.charge < 0][dihad_evts].flatten()),
)
# make Had/Lep decaying objects for SEMILEP events
top_isLep = (GenTop.decaytype == 1)[semilep_evts].flatten()
Lep_Top = JaggedCandidateArray.candidatesfromcounts(
counts=semilep_evts.astype(int),
pt=np.where(top_isLep, GenTop[semilep_evts].pt.flatten(),
GenTbar[semilep_evts].pt.flatten()),
eta=np.where(top_isLep, GenTop[semilep_evts].eta.flatten(),
GenTbar[semilep_evts].eta.flatten()),
phi=np.where(top_isLep, GenTop[semilep_evts].phi.flatten(),
GenTbar[semilep_evts].phi.flatten()),
mass=np.where(top_isLep, GenTop[semilep_evts].mass.flatten(),
GenTbar[semilep_evts].mass.flatten()),
charge=np.where(top_isLep, GenTop[semilep_evts].charge.flatten(),
GenTbar[semilep_evts].charge.flatten()),
decaytype=np.where(top_isLep, GenTop[semilep_evts].decaytype.flatten(),
GenTbar[semilep_evts].decaytype.flatten()),
)
Had_Top = JaggedCandidateArray.candidatesfromcounts(
counts=semilep_evts.astype(int),
pt=np.where(~top_isLep, GenTop[semilep_evts].pt.flatten(),
GenTbar[semilep_evts].pt.flatten()),
eta=np.where(~top_isLep, GenTop[semilep_evts].eta.flatten(),
GenTbar[semilep_evts].eta.flatten()),
phi=np.where(~top_isLep, GenTop[semilep_evts].phi.flatten(),
GenTbar[semilep_evts].phi.flatten()),
mass=np.where(~top_isLep, GenTop[semilep_evts].mass.flatten(),
GenTbar[semilep_evts].mass.flatten()),
charge=np.where(~top_isLep, GenTop[semilep_evts].charge.flatten(),
GenTbar[semilep_evts].charge.flatten()),
decaytype=np.where(~top_isLep,
GenTop[semilep_evts].decaytype.flatten(),
GenTbar[semilep_evts].decaytype.flatten()),
)
Lep_B = JaggedCandidateArray.candidatesfromcounts(
counts=semilep_evts.astype(int),
pt=np.where(top_isLep, GenB[semilep_evts].pt.flatten(),
GenBbar[semilep_evts].pt.flatten()),
eta=np.where(top_isLep, GenB[semilep_evts].eta.flatten(),
GenBbar[semilep_evts].eta.flatten()),
phi=np.where(top_isLep, GenB[semilep_evts].phi.flatten(),
GenBbar[semilep_evts].phi.flatten()),
mass=np.where(top_isLep, GenB[semilep_evts].mass.flatten(),
GenBbar[semilep_evts].mass.flatten()),
charge=np.where(top_isLep, GenB[semilep_evts].charge.flatten(),
GenBbar[semilep_evts].charge.flatten()),
decaytype=np.where(top_isLep, GenB[semilep_evts].decaytype.flatten(),
GenBbar[semilep_evts].decaytype.flatten()),
)
Had_B = JaggedCandidateArray.candidatesfromcounts(
counts=semilep_evts.astype(int),
pt=np.where(~top_isLep, GenB[semilep_evts].pt.flatten(),
GenBbar[semilep_evts].pt.flatten()),
eta=np.where(~top_isLep, GenB[semilep_evts].eta.flatten(),
GenBbar[semilep_evts].eta.flatten()),
phi=np.where(~top_isLep, GenB[semilep_evts].phi.flatten(),
GenBbar[semilep_evts].phi.flatten()),
mass=np.where(~top_isLep, GenB[semilep_evts].mass.flatten(),
GenBbar[semilep_evts].mass.flatten()),
charge=np.where(~top_isLep, GenB[semilep_evts].charge.flatten(),
GenBbar[semilep_evts].charge.flatten()),
decaytype=np.where(~top_isLep, GenB[semilep_evts].decaytype.flatten(),
GenBbar[semilep_evts].decaytype.flatten()),
)
SEMILEP_evts = awkward.Table(
TTbar=awkward.JaggedArray.fromcounts(semilep_evts.astype(int),
GenTTbar[semilep_evts].flatten()),
THad=awkward.JaggedArray.fromcounts(semilep_evts.astype(int),
Had_Top.flatten()),
TLep=awkward.JaggedArray.fromcounts(semilep_evts.astype(int),
Lep_Top.flatten()),
BHad=awkward.JaggedArray.fromcounts(semilep_evts.astype(int),
Had_B.flatten()),
BLep=awkward.JaggedArray.fromcounts(semilep_evts.astype(int),
Lep_B.flatten()),
WHad=awkward.JaggedArray.fromcounts(
semilep_evts.astype(int),
wbosons[wbosons.decaytype == 2][semilep_evts].flatten()),
WLep=awkward.JaggedArray.fromcounts(
semilep_evts.astype(int),
wbosons[wbosons.decaytype == 1][semilep_evts].flatten()),
Lepton=awkward.JaggedArray.fromcounts(
semilep_evts.astype(int), charged_leps[semilep_evts].flatten()),
Nu=awkward.JaggedArray.fromcounts(
semilep_evts.astype(int), neutral_leps[semilep_evts].flatten()),
WJa=awkward.JaggedArray.fromcounts(semilep_evts.astype(int),
First[semilep_evts].flatten()),
WJb=awkward.JaggedArray.fromcounts(semilep_evts.astype(int),
Second[semilep_evts].flatten()),
Up_Had=awkward.JaggedArray.fromcounts(
semilep_evts.astype(int), wpartons_up[semilep_evts].flatten()),
Down_Had=awkward.JaggedArray.fromcounts(
semilep_evts.astype(int), wpartons_dw[semilep_evts].flatten()),
)
GenTTbar = awkward.Table(
SL=SEMILEP_evts,
DL=DILEP_evts,
Had=HAD_evts,
)
return GenTTbar
def candidatesfromoffsets(cls, offsets, **kwargs):
"""
cands = JaggedCandidateArray.candidatesfromoffsets(offsets=offsets,
pt=column1,
eta=column2,
phi=column3,
mass=column4,
...)
"""
items = kwargs
argkeys = items.keys()
p4 = None
fast_pt = None
fast_eta = None
fast_phi = None
fast_mass = None
if 'p4' in argkeys:
p4 = items['p4']
if not isinstance(p4, uproot_methods.TLorentzVectorArray):
p4 = uproot_methods.TLorentzVectorArray.from_cartesian(
p4[:, 0], p4[:, 1], p4[:, 2], p4[:, 3])
fast_pt = _fast_pt(p4)
fast_eta = _fast_eta(p4)
fast_phi = _fast_phi(p4)
fast_mass = _fast_mass(p4)
elif 'pt' in argkeys and 'eta' in argkeys and 'phi' in argkeys and 'mass' in argkeys:
p4 = uproot_methods.TLorentzVectorArray.from_ptetaphim(
items['pt'], items['eta'], items['phi'], items['mass'])
fast_pt = items['pt']
fast_eta = items['eta']
fast_phi = items['phi']
fast_mass = items['mass']
del items['pt']
del items['eta']
del items['phi']
del items['mass']
elif 'pt' in argkeys and 'eta' in argkeys and 'phi' in argkeys and 'energy' in argkeys:
p4 = uproot_methods.TLorentzVectorArray.from_ptetaphi(
items['pt'], items['eta'], items['phi'], items['energy'])
fast_pt = items['pt']
fast_eta = items['eta']
fast_phi = items['phi']
fast_mass = _fast_mass(p4)
del items['pt']
del items['eta']
del items['phi']
del items['energy']
elif 'px' in argkeys and 'py' in argkeys and 'pz' in argkeys and 'mass' in argkeys:
p4 = uproot_methods.TLorentzVectorArray.from_xyzm(
items['px'], items['py'], items['pz'], items['mass'])
fast_pt = _fast_pt(p4)
fast_eta = _fast_eta(p4)
fast_phi = _fast_phi(p4)
fast_mass = items['mass']
del items['px']
del items['py']
del items['pz']
del items['mass']
elif 'pt' in argkeys and 'phi' in argkeys and 'pz' in argkeys and 'energy' in argkeys:
p4 = uproot_methods.TLorentzVectorArray.from_cylindrical(
items['pt'], items['phi'], items['pz'], items['energy'])
fast_pt = items['pt']
fast_eta = _fast_eta(p4)
fast_phi = items['phi']
fast_mass = _fast_mass(p4)
del items['pt']
del items['phi']
del items['pz']
del items['energy']
elif 'px' in argkeys and 'py' in argkeys and 'pz' in argkeys and 'energy' in argkeys:
p4 = uproot_methods.TLorentzVectorArray.from_cartesian(
items['px'], items['py'], items['pz'], items['energy'])
fast_pt = _fast_pt(p4)
fast_eta = _fast_eta(p4)
fast_phi = _fast_phi(p4)
fast_mass = _fast_mass(p4)
del items['px']
del items['py']
del items['pz']
del items['energy']
elif 'p' in argkeys and 'theta' in argkeys and 'phi' in argkeys and 'energy' in argkeys:
p4 = uproot_methods.TLorentzVectorArray.from_spherical(
items['p'], items['theta'], items['phi'], items['energy'])
fast_pt = _fast_pt(p4)
fast_eta = _fast_eta(p4)
fast_phi = items['phi']
fast_mass = _fast_mass(p4)
del items['p']
del items['theta']
del items['phi']
del items['energy']
elif 'p3' in argkeys and 'energy' in argkeys:
p4 = uproot_methods.TLorentzVectorArray.from_p3(
items['p3'], items['energy'])
fast_pt = _fast_pt(p4)
fast_eta = _fast_eta(p4)
fast_phi = _fast_phi(p4)
fast_mass = _fast_mass(p4)
del items['p3']
del items['energy']
else:
raise Exception(
'No valid definition of four-momentum found to build JaggedCandidateArray'
)
items['p4'] = p4
items['__fast_pt'] = fast_pt
items['__fast_eta'] = fast_eta
items['__fast_phi'] = fast_phi
items['__fast_mass'] = fast_mass
return cls.fromoffsets(offsets, awkward.Table(items))
def arrow_table(self):
for object_array in self.event_iterator.iterate():
unicode_array = {}
for key in object_array.keys():
unicode_array[key.decode('UTF8')] = object_array[key]
yield awkward.toarrow(awkward.Table(unicode_array))
def run_query(input_filenames=None, tree_name=None):
import awkward, uproot
return (lambda event: (awkward.Table if hasattr(awkward, 'Table') else awkward['Table'])((event.MVA3lCERN_weight_ttH if hasattr(event, 'MVA3lCERN_weight_ttH') else event['MVA3lCERN_weight_ttH'])))((lambda event: event[(((((((((((((((((((((((((((event.trilep_type if hasattr(event, 'trilep_type') else event['trilep_type']) > 0) & ((event.nTaus_OR_Pt25 if hasattr(event, 'nTaus_OR_Pt25') else event['nTaus_OR_Pt25']) == 0)) & (abs((event.total_charge if hasattr(event, 'total_charge') else event['total_charge'])) == 1)) & ((event.nJets_OR_T if hasattr(event, 'nJets_OR_T') else event['nJets_OR_T']) >= 2)) & ((event.nJets_OR_T_MV2c10_70 if hasattr(event, 'nJets_OR_T_MV2c10_70') else event['nJets_OR_T_MV2c10_70']) > 0)) & ((event.lep_Pt_1 if hasattr(event, 'lep_Pt_1') else event['lep_Pt_1']) > 15000.0)) & ((event.lep_Pt_2 if hasattr(event, 'lep_Pt_2') else event['lep_Pt_2']) > 15000.0)) & ((event.lep_isolationFixedCutLoose_0 if hasattr(event, 'lep_isolationFixedCutLoose_0') else event['lep_isolationFixedCutLoose_0']) > 0)) & (abs(((event.Mlll012 if hasattr(event, 'Mlll012') else event['Mlll012']) - 91200.0)) > 10000.0)) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) != (-(event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1']))) | ((event.Mll01 if hasattr(event, 'Mll01') else event['Mll01']) > 12000.0))) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) != (-(event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2']))) | ((event.Mll02 if hasattr(event, 'Mll02') else event['Mll02']) > 12000.0))) & (((((abs((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0'])) == 13) & ((event.lep_isMedium_0 if hasattr(event, 'lep_isMedium_0') else event['lep_isMedium_0']) > 0)) | (abs((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0'])) == 11)) & (((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) == 11) & (abs((event.lep_Eta_1 if hasattr(event, 'lep_Eta_1') else event['lep_Eta_1'])) < 2.0)) | ((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) == 13) & ((event.lep_isMedium_1 if hasattr(event, 'lep_isMedium_1') else event['lep_isMedium_1']) > 0)))) & (((abs((event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2'])) == 11) & (abs((event.lep_Eta_2 if hasattr(event, 'lep_Eta_2') else event['lep_Eta_2'])) < 2.0)) | ((abs((event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2'])) == 13) & ((event.lep_isMedium_2 if hasattr(event, 'lep_isMedium_2') else event['lep_isMedium_2']) > 0))))) & ((((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) * abs((event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2']))) != 169) & ((event.DRll12 if hasattr(event, 'DRll12') else event['DRll12']) > 0.5)) | ((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) * abs((event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2']))) == 169))) & ((event.Mll12 if hasattr(event, 'Mll12') else event['Mll12']) > 12000.0)) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) != (-(event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1']))) | (abs(((event.Mll01 if hasattr(event, 'Mll01') else event['Mll01']) - 91200.0)) > 10000.0))) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) != (-(event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2']))) | (abs(((event.Mll02 if hasattr(event, 'Mll02') else event['Mll02']) - 91200.0)) > 10000.0))) & ((event.MVA3lCERN_weight_ttH if hasattr(event, 'MVA3lCERN_weight_ttH') else event['MVA3lCERN_weight_ttH']) > (-1))) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) != (-(event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1']))) | (abs(((event.Mll01 if hasattr(event, 'Mll01') else event['Mll01']) - 91200.0)) > 10000.0))) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) != (-(event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2']))) | (abs(((event.Mll02 if hasattr(event, 'Mll02') else event['Mll02']) - 91200.0)) > 10000.0))) & ((event.MVA3lCERN_weight_ttH if hasattr(event, 'MVA3lCERN_weight_ttH') else event['MVA3lCERN_weight_ttH']) > 0.3)) & ((event.MVA3lCERN_weight_ttW if hasattr(event, 'MVA3lCERN_weight_ttW') else event['MVA3lCERN_weight_ttW']) < 0.75)) & ((event.MVA3lCERN_weight_ttZ if hasattr(event, 'MVA3lCERN_weight_ttZ') else event['MVA3lCERN_weight_ttZ']) < 0.75)) & ((event.MVA3lCERN_weight_VV if hasattr(event, 'MVA3lCERN_weight_VV') else event['MVA3lCERN_weight_VV']) < 0.75)) & ((event.MVA3lCERN_weight_ttbar if hasattr(event, 'MVA3lCERN_weight_ttbar') else event['MVA3lCERN_weight_ttbar']) < 0.3)) & ((((((((event.dilep_type if hasattr(event, 'dilep_type') else event['dilep_type']) > 0) & (((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0']) * (event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) > 0)) & ((event.lep_isQMisID_1 if hasattr(event, 'lep_isQMisID_1') else event['lep_isQMisID_1']) == 0)) & ((event.lep_isQMisID_0 if hasattr(event, 'lep_isQMisID_0') else event['lep_isQMisID_0']) == 0)) | ((((event.trilep_type if hasattr(event, 'trilep_type') else event['trilep_type']) > 0) & ((event.lep_isQMisID_2 if hasattr(event, 'lep_isQMisID_2') else event['lep_isQMisID_2']) == 0)) & ((event.lep_isQMisID_1 if hasattr(event, 'lep_isQMisID_1') else event['lep_isQMisID_1']) == 0))) | (((event.quadlep_type if hasattr(event, 'quadlep_type') else event['quadlep_type']) > 0) & ((event.FSF_4L_tot if hasattr(event, 'FSF_4L_tot') else event['FSF_4L_tot']) == 1))) & (((((((((event.dilep_type if hasattr(event, 'dilep_type') else event['dilep_type']) > 0) & (((((abs((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0'])) == 13) & ((event.lep_isMedium_0 if hasattr(event, 'lep_isMedium_0') else event['lep_isMedium_0']) > 0)) & ((event.lep_isolationFixedCutLoose_0 if hasattr(event, 'lep_isolationFixedCutLoose_0') else event['lep_isolationFixedCutLoose_0']) > 0)) & ((event.lep_promptLeptonVeto_TagWeight_0 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_0') else event['lep_promptLeptonVeto_TagWeight_0']) < (-0.5))) | ((((((abs((event.lep_ID_0 if hasattr(event, 'lep_ID_0') else event['lep_ID_0'])) == 11) & ((event.lep_isolationFixedCutLoose_0 if hasattr(event, 'lep_isolationFixedCutLoose_0') else event['lep_isolationFixedCutLoose_0']) > 0)) & ((event.lep_isTightLH_0 if hasattr(event, 'lep_isTightLH_0') else event['lep_isTightLH_0']) > 0)) & ((event.lep_chargeIDBDTTight_0 if hasattr(event, 'lep_chargeIDBDTTight_0') else event['lep_chargeIDBDTTight_0']) > 0.7)) & ((event.lep_ambiguityType_0 if hasattr(event, 'lep_ambiguityType_0') else event['lep_ambiguityType_0']) == 0)) & ((event.lep_promptLeptonVeto_TagWeight_0 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_0') else event['lep_promptLeptonVeto_TagWeight_0']) < (-0.7))))) & (((((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) == 13) & ((event.lep_isMedium_1 if hasattr(event, 'lep_isMedium_1') else event['lep_isMedium_1']) > 0)) & ((event.lep_isolationFixedCutLoose_1 if hasattr(event, 'lep_isolationFixedCutLoose_1') else event['lep_isolationFixedCutLoose_1']) > 0)) & ((event.lep_promptLeptonVeto_TagWeight_1 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_1') else event['lep_promptLeptonVeto_TagWeight_1']) < (-0.5))) | ((((((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) == 11) & ((event.lep_isolationFixedCutLoose_1 if hasattr(event, 'lep_isolationFixedCutLoose_1') else event['lep_isolationFixedCutLoose_1']) > 0)) & ((event.lep_isTightLH_1 if hasattr(event, 'lep_isTightLH_1') else event['lep_isTightLH_1']) > 0)) & ((event.lep_chargeIDBDTTight_1 if hasattr(event, 'lep_chargeIDBDTTight_1') else event['lep_chargeIDBDTTight_1']) > 0.7)) & ((event.lep_ambiguityType_1 if hasattr(event, 'lep_ambiguityType_1') else event['lep_ambiguityType_1']) == 0)) & ((event.lep_promptLeptonVeto_TagWeight_1 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_1') else event['lep_promptLeptonVeto_TagWeight_1']) < (-0.7))))) | ((((event.trilep_type if hasattr(event, 'trilep_type') else event['trilep_type']) > 0) & ((event.nTaus_OR_Pt25 if hasattr(event, 'nTaus_OR_Pt25') else event['nTaus_OR_Pt25']) == 0)) & (((((abs((event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2'])) == 13) & ((event.lep_isolationFixedCutLoose_2 if hasattr(event, 'lep_isolationFixedCutLoose_2') else event['lep_isolationFixedCutLoose_2']) > 0)) & ((event.lep_promptLeptonVeto_TagWeight_2 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_2') else event['lep_promptLeptonVeto_TagWeight_2']) < (-0.5))) | ((((((abs((event.lep_ID_2 if hasattr(event, 'lep_ID_2') else event['lep_ID_2'])) == 11) & ((event.lep_isolationFixedCutLoose_2 if hasattr(event, 'lep_isolationFixedCutLoose_2') else event['lep_isolationFixedCutLoose_2']) > 0)) & ((event.lep_isTightLH_2 if hasattr(event, 'lep_isTightLH_2') else event['lep_isTightLH_2']) > 0)) & ((event.lep_chargeIDBDTTight_2 if hasattr(event, 'lep_chargeIDBDTTight_2') else event['lep_chargeIDBDTTight_2']) > 0.7)) & ((event.lep_promptLeptonVeto_TagWeight_2 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_2') else event['lep_promptLeptonVeto_TagWeight_2']) < (-0.7))) & ((event.lep_ambiguityType_2 if hasattr(event, 'lep_ambiguityType_2') else event['lep_ambiguityType_2']) == 0))) & ((((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) == 13) & ((event.lep_isolationFixedCutLoose_1 if hasattr(event, 'lep_isolationFixedCutLoose_1') else event['lep_isolationFixedCutLoose_1']) > 0)) & ((event.lep_promptLeptonVeto_TagWeight_1 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_1') else event['lep_promptLeptonVeto_TagWeight_1']) < (-0.5))) | ((((((abs((event.lep_ID_1 if hasattr(event, 'lep_ID_1') else event['lep_ID_1'])) == 11) & ((event.lep_isolationFixedCutLoose_1 if hasattr(event, 'lep_isolationFixedCutLoose_1') else event['lep_isolationFixedCutLoose_1']) > 0)) & ((event.lep_isTightLH_1 if hasattr(event, 'lep_isTightLH_1') else event['lep_isTightLH_1']) > 0)) & ((event.lep_chargeIDBDTTight_1 if hasattr(event, 'lep_chargeIDBDTTight_1') else event['lep_chargeIDBDTTight_1']) > 0.7)) & ((event.lep_promptLeptonVeto_TagWeight_1 if hasattr(event, 'lep_promptLeptonVeto_TagWeight_1') else event['lep_promptLeptonVeto_TagWeight_1']) < (-0.7))) & ((event.lep_ambiguityType_1 if hasattr(event, 'lep_ambiguityType_1') else event['lep_ambiguityType_1']) == 0)))))) | (((event.dilep_type if hasattr(event, 'dilep_type') else event['dilep_type']) > 0) & ((event.nTaus_OR_Pt25 if hasattr(event, 'nTaus_OR_Pt25') else event['nTaus_OR_Pt25']) > 1))) | ((((event.dilep_type if hasattr(event, 'dilep_type') else event['dilep_type']) > 0) | ((event.trilep_type if hasattr(event, 'trilep_type') else event['trilep_type']) > 0)) == 0)) | ((event.quadlep_type if hasattr(event, 'quadlep_type') else event['quadlep_type']) > 0)) | (((event.trilep_type if hasattr(event, 'trilep_type') else event['trilep_type']) > 0) & ((event.nTaus_OR_Pt25 if hasattr(event, 'nTaus_OR_Pt25') else event['nTaus_OR_Pt25']) > 0))))) | ((event.is1L2Tau if hasattr(event, 'is1L2Tau') else event['is1L2Tau']) > 0))])((lambda input_files: uproot.lazyarrays(input_files, tree_name if tree_name is not None else (lambda key_array: key_array[key_array[:, 1] == 'TTree'][:, 0])(awkward.Table(uproot.open(input_files[0]).classnames()).unzip()[0])[0]))(input_filenames)))
