-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathmutau_bkgd_skim.py
266 lines (226 loc) · 11.7 KB
/
mutau_bkgd_skim.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
import sys
import uproot
import argparse
import json
import pyarrow as pa
import pyarrow.parquet as pq
import numpy as np
from coffea import processor
from coffea.nanoevents import NanoEventsFactory, PFNanoAODSchema, NanoAODSchema
from coffea.dataset_tools import (
apply_to_fileset,
max_chunks,
preprocess,
)
from coffea.lumi_tools import LumiData, LumiList, LumiMask
import awkward as ak
import dask
from dask import config as cfg
cfg.set({'distributed.scheduler.worker-ttl': None}) # Check if this solves some dask issues
import dask_awkward as dak
# from dask_jobqueue import HTCondorCluster
from dask.distributed import Client, wait, progress, LocalCluster
from fileset import *
import ROOT
import warnings
warnings.filterwarnings("ignore", module="coffea") # Suppress annoying deprecation warnings for coffea vector, c.f. https://github.com/CoffeaTeam/coffea/blob/master/src/coffea/nanoevents/methods/candidate.py
import logging
# Can also be put in a utils file later
def delta_r_mask(first: ak.highlevel.Array, second: ak.highlevel.Array, threshold: float) -> ak.highlevel.Array:
mval = first.metric_table(second)
return ak.all(mval > threshold, axis=-1)
PFNanoAODSchema.mixins["DisMuon"] = "Muon"
def main():
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
client = Client()
print(f'Dask dashboard available at: {client.dashboard_link}')
class MyProcessor(processor.ProcessorABC):
def __init__(self):
pass
def process(self, events):
# Determine if dataset is MC or Data
is_MC = True if hasattr(events, "GenPart") else False
if is_MC: sumWeights = ak.sum(events.genWeight)
logger.info("Starting process")
lumi = 0
if not is_MC:
lumi_mask = LumiMask("./tools/Cert_314472-325175_13TeV_Legacy2018_Collisions18_JSON.txt")
lumi_data = LumiData("./tools/LumiData_2018_20200401.csv", is_inst_lumi=True)
events = events[lumi_mask(events.run, events.luminosityBlock)]
lumi_list = LumiList(*dask.compute(events.run, events.luminosityBlock))
lumi = lumi_data.get_lumi(lumi_list)/10**9 # convert from inverse microbarn to inverse femtobarn
charged_sel = events.Jet.constituents.pf.charge != 0
events["Jet_dxy"] = dak.flatten(ak.drop_none(events.Jet.constituents.pf[ak.argmax(events.Jet.constituents.pf[charged_sel].pt, axis=2, keepdims=True)].d0), axis=-1)
# Define the "good muon" condition for each muon per event
good_muon_mask = (
#((events.DisMuon.isGlobal == 1) | (events.DisMuon.isTracker == 1)) & # Equivalent to loose ID cut without isPFcand requirement. Reduce background from non-prompt muons
(events.DisMuon.pt > 20)
& (abs(events.DisMuon.eta) < 2.4) # Acceptance of the CMS muon system
)
logger.info("Defined good muons")
events['DisMuon'] = ak.drop_none(events.DisMuon[good_muon_mask])
logger.info("Applied mask to DisMuon")
num_evts = ak.num(events, axis=0)
logger.info("Counted the number of original events")
num_good_muons = ak.count_nonzero(good_muon_mask, axis=1)
logger.info("Counted the number of events with good muons")
events = events[num_good_muons >= 1]
logger.info("Counted the number of events with one or more good muons")
logger.info("Cut muons")
# Perform the overlap removal with respect to muons, electrons and photons, dR=0.4
events['Jet'] = events.Jet[delta_r_mask(events.Jet, events.Photon, 0.4)]
events['Jet'] = events.Jet[delta_r_mask(events.Jet, events.Electron, 0.4)]
events['Jet'] = events.Jet[delta_r_mask(events.Jet, events.Muon, 0.4)]
events['Jet'] = events.Jet[delta_r_mask(events.Jet, events.DisMuon, 0.4)]
logger.info("Performed overlap removal")
good_jet_mask = (
(events.Jet.pt > 20)
& (abs(events.Jet.eta) < 2.4)
)
logger.info("Defined good jets")
events['Jet'] = events.Jet[good_jet_mask]
num_good_jets = ak.count_nonzero(good_jet_mask, axis=1)
events = events[num_good_jets >= 1]
logger.info("Cut jets")
#Noise filter
noise_mask = (
(events.Flag.goodVertices == 1)
& (events.Flag.globalSuperTightHalo2016Filter == 1)
& (events.Flag.EcalDeadCellTriggerPrimitiveFilter == 1)
& (events.Flag.BadPFMuonFilter == 1)
& (events.Flag.BadPFMuonDzFilter == 1)
& (events.Flag.hfNoisyHitsFilter == 1)
& (events.Flag.eeBadScFilter == 1)
& (events.Flag.ecalBadCalibFilter == 1)
)
events = events[noise_mask]
#Trigger Selection
trigger_mask = (
events.HLT.PFMET120_PFMHT120_IDTight |\
events.HLT.PFMET130_PFMHT130_IDTight |\
events.HLT.PFMET140_PFMHT140_IDTight |\
events.HLT.PFMETNoMu120_PFMHTNoMu120_IDTight |\
events.HLT.PFMETNoMu130_PFMHTNoMu130_IDTight |\
events.HLT.PFMETNoMu140_PFMHTNoMu140_IDTight |\
events.HLT.PFMET120_PFMHT120_IDTight_PFHT60 |\
#events.HLT.MonoCentralPFJet80_PFMETNoMu120_PFMHTNoMu120_IDTight |\
events.HLT.PFMETTypeOne140_PFMHT140_IDTight |\
events.HLT.MET105_IsoTrk50 |\
events.HLT.PFMETNoMu110_PFMHTNoMu110_IDTight_FilterHF |\
events.HLT.MET120_IsoTrk50 |\
events.HLT.IsoMu24_eta2p1_MediumDeepTauPFTauHPS35_L2NN_eta2p1_CrossL1 |\
events.HLT.IsoMu24_eta2p1_MediumDeepTauPFTauHPS30_L2NN_eta2p1_CrossL1 |\
events.HLT.Ele30_WPTight_Gsf |\
events.HLT.DoubleMediumDeepTauPFTauHPS35_L2NN_eta2p1 |\
events.HLT.DoubleMediumChargedIsoDisplacedPFTauHPS32_Trk1_eta2p1 |\
events.HLT.DoubleMediumChargedIsoPFTauHPS40_Trk1_eta2p1
)
events = events[trigger_mask]
meta = ak.Array([0], backend = "typetracer")
event_counts = events.map_partitions(lambda part: ak.num(part, axis = 0), meta = meta)
partition_counts = event_counts.compute()
non_empty_partitions = [
events.partitions[i] for i in range(len(partition_counts)) if partition_counts[i] > 0
]
if non_empty_partitions:
events = dak.concatenate(non_empty_partitions)
if is_MC: weights = events.genWeight / sumWeights
else: weights = ak.ones_like(events.event) # Classic move to create a 1d-array of ones, the appropriate weight for data
logger.info("Defined weights")
out_dict = {}
muon_vars = [
"pt",
"eta",
"phi",
"charge",
"dxy",
"dxyErr",
"dz",
"dzErr",
"looseId",
"mediumId",
"tightId",
"pfRelIso03_all",
"pfRelIso03_chg",
"pfRelIso04_all",
"tkRelIso",
]
jet_vars = [
"pt",
"eta",
"phi",
"disTauTag_score1",
]
gpart_vars = [
"genPartIdxMother",
"statusFlags",
"pdgId",
"status",
"eta",
"mass",
"phi",
"pt",
"vertexR",
"vertexRho",
"vx",
"vy",
"vz",
]
gvist_vars = [
"genPartIdxMother",
"charge",
"status",
"eta",
"mass",
"phi",
"pt",
]
tau_vars = events.Tau.fields
MET_vars = events.PFMET.fields
for branch in muon_vars:
out_dict["DisMuon_" + branch] = ak.drop_none(events["DisMuon"][branch])
for branch in jet_vars:
out_dict["Jet_" + branch] = ak.drop_none(events["Jet"][branch])
for branch in gpart_vars:
out_dict["GenPart_" + branch] = ak.drop_none(events["GenPart"][branch])
for branch in gvist_vars:
out_dict["GenVisTau_" + branch] = ak.drop_none(events["GenVisTau"][branch])
for branch in tau_vars:
out_dict["Tau_" + branch] = ak.drop_none(events["Tau"][branch])
for branch in MET_vars:
out_dict["PFMET_" + branch] = ak.drop_none(events["PFMET"][branch])
out_dict["event"] = ak.drop_none(events.event)
out_dict["run"] = ak.drop_none(events.run)
out_dict["luminosityBlock"] = ak.drop_none(events.luminosityBlock)
out_dict["Jet_dxy"] = ak.drop_none(events["Jet_dxy"])
out_dict["nDisMuon"] = dak.num(events.DisMuon)
out_dict["nJet"] = dak.num(events.Jet)
out_dict["nGenPart"] = dak.num(events.GenPart)
out_dict["nGenVisTau"] = dak.num(events.GenVisTau)
out_dict["nTau"] = dak.num(events.Tau)
logger.info(f"Filled dictionary")
out_dict = dak.zip(out_dict, depth_limit = 1)
logger.info(f"Dictionary zipped")
return uproot.dask_write(out_dict, "my_skim_muon_" + events.metadata['dataset'] + "_root", tree_name="Events")
logger.info("Dictionary written to root files")
def postprocess(self, accumulator):
pass
dataset_runnable, dataset_updated = preprocess(
fileset,
align_clusters=False,
step_size=100_00,
files_per_batch=1,
skip_bad_files=True,
save_form=False,
)
to_compute = apply_to_fileset(
MyProcessor(),
max_chunks(dataset_runnable, 1000000000),
schemaclass=PFNanoAODSchema
)
(out,) = dask.compute(to_compute)
MyProcessor().postprocess(out)
print(out)
if __name__ == "__main__":
main()