def main():
# Specify some output dirs
out_dir_private_UL = os.path.join(topcoffea_path("json"),
"signal_samples/private_UL/")
out_dir_top19001_local = os.path.join(
topcoffea_path("json"), "signal_samples/private_top19001_local")
out_dir_central_UL = os.path.join(topcoffea_path("json"),
"signal_samples/central_UL/")
out_dir_central_2017 = os.path.join(topcoffea_path("json"),
"signal_samples/central_2017/")
# Private UL
#make_jsons_for_dict_of_samples(private_UL17_dict,"/hadoop","2017",out_dir_private_UL)
#make_jsons_for_dict_of_samples(private_UL18_dict,"/hadoop","2018",out_dir_private_UL)
#make_jsons_for_dict_of_samples(private_UL16_dict,"/hadoop","2016",out_dir_private_UL)
#make_jsons_for_dict_of_samples(private_UL16APV_dict,"/hadoop","2016APV",out_dir_private_UL) # Not sure what we need here for the year, can remake the JSONs later to update when we have SFs etc set up for 2016 stuff (right now I think it's mostly just 18)
# TOP-19-001 ttll
#make_jsons_for_dict_of_samples(private_2017_dict,"","2017",out_dir_top19001_local)
# Central
#make_jsons_for_dict_of_samples(central_2017_correctnPartonsInBorn_dict,"root://ndcms.crc.nd.edu/","2017",out_dir_central_2017,on_das=True)
make_jsons_for_dict_of_samples(central_2017_dict,
"root://ndcms.crc.nd.edu/",
"2017",
out_dir_central_2017,
on_das=True)
def GetBTagSF(eta, pt, flavor, year=2018, sys=0):
# Efficiencies and SFs for UL only available for 2017 and 2018
if year == 2016: SFevaluatorBtag = BTagScaleFactor(topcoffea_path("data/btagSF/DeepFlav_2016.csv"),"MEDIUM")
elif year == 2017: SFevaluatorBtag = BTagScaleFactor(topcoffea_path("data/btagSF/UL/DeepJet_UL17.csv"),"MEDIUM")
elif year == 2018: SFevaluatorBtag = BTagScaleFactor(topcoffea_path("data/btagSF/UL/DeepJet_UL18.csv"),"MEDIUM")
if sys==0 : SF=SFevaluatorBtag.eval("central",flavor,eta,pt)
elif sys==1 : SF=SFevaluatorBtag.eval("up",flavor,eta,pt)
elif sys==-1: SF=SFevaluatorBtag.eval("down",flavor,eta,pt)
return (SF)
def GetMCeffFunc(WP='medium', flav='b', year=2018):
pathToBtagMCeff = topcoffea_path('data/btagSF/UL/btagMCeff_%i.pkl.gz' %
year)
hists = {}
with gzip.open(pathToBtagMCeff) as fin:
hin = pickle.load(fin)
for k in hin.keys():
if k in hists: hists[k] += hin[k]
else: hists[k] = hin[k]
h = hists['jetptetaflav']
hnum = h.integrate('WP', WP)
hden = h.integrate('WP', 'all')
getnum = lookup_tools.dense_lookup.dense_lookup(
hnum.values(overflow='over')[()], [
hnum.axis('pt').edges(),
hnum.axis('abseta').edges(),
hnum.axis('flav').edges()
])
getden = lookup_tools.dense_lookup.dense_lookup(
hden.values(overflow='over')[()], [
hden.axis('pt').edges(),
hnum.axis('abseta').edges(),
hden.axis('flav').edges()
])
values = hnum.values(overflow='over')[()]
edges = [
hnum.axis('pt').edges(),
hnum.axis('abseta').edges(),
hnum.axis('flav').edges()
]
fun = lambda pt, abseta, flav: getnum(pt, abseta, flav) / getden(
pt, abseta, flav)
return fun
def main():
# Specify some output dirs
out_dir_private_UL = os.path.join(topcoffea_path("json"),
"signal_samples/private_UL/")
out_dir_top19001_local = os.path.join(
topcoffea_path("json"), "signal_samples/private_top19001_local")
out_dir_central_UL = os.path.join(topcoffea_path("json"),
"signal_samples/central_UL/")
out_dir_central_2017 = os.path.join(topcoffea_path("json"),
"signal_samples/central_2017/")
# Private UL
#make_jsons_for_dict_of_samples(private_UL17_dict,"/hadoop","2017",out_dir_private_UL)
#make_jsons_for_dict_of_samples(private_UL18_dict,"/hadoop","2018",out_dir_private_UL)
#make_jsons_for_dict_of_samples(private_UL16_dict,"/hadoop","2016",out_dir_private_UL)
make_jsons_for_dict_of_samples(
private_UL16APV_dict, "/hadoop", "2016APV", out_dir_private_UL
) # Not sure what we need here for the year, can remake the JSONs later to update when we have SFs etc set up for 2016 stuff (right now I think it's mostly just 18)
def __init__(self):
# The path to the file listing the luminosity for each year
self.LUMI_FILE = topcoffea_path("json/lumi.json")
# The order of the categories in the TOP-19-001 AN yield tables
self.CAT_LST = [
"cat_2lss_p", "cat_2lss_m", "cat_3l_1b_offZ_p", "cat_3l_1b_offZ_m",
"cat_3l_2b_offZ_p", "cat_3l_2b_offZ_m", "cat_3l_1b_onZ",
"cat_3l_2b_onZ", "cat_4l"
]
# A dictionary mapping names of samples in the samples axis to a short version of the name
self.PROC_MAP = {
"ttlnu": [
"ttW_centralUL17", "ttlnu_private2017", "ttlnuJet_privateUL17",
"ttlnuJet_privateUL18"
],
"ttll": [
"ttZ_centralUL17", "ttll_TOP-19-001", "ttllJet_privateUL17",
"ttllJet_privateUL18"
],
"ttH": [
"ttH_centralUL17", "ttH_private2017", "ttHJet_privateUL17",
"ttHJet_privateUL18"
],
"tllq": [
"tZq_centralUL17", "tllq_private2017", "tllq_privateUL17",
"tllq_privateUL18"
],
"tHq": ["tHq_central2017", "tHq_privateUL17"],
"tttt": ["tttt_central2017", "tttt_privateUL17"],
}
# The jet bins we define for the lep categories
self.JET_BINS = {
"2lss": [4, 5, 6, 7],
"3l": [2, 3, 4, 5],
"4l": [2, 3, 4],
}
# The sub categories of the lep categories
self.ch_3l_onZ = ["eemSSonZ", "mmeSSonZ", "eeeSSonZ", "mmmSSonZ"]
self.ch_3l_offZ = ["eemSSoffZ", "mmeSSoffZ", "eeeSSoffZ", "mmmSSoffZ"]
self.ch_2lss = ["eeSSonZ", "eeSSoffZ", "mmSSonZ", "mmSSoffZ", "emSS"]
self.ch_4l = ["eeee", "eeem", "eemm", "mmme", "mmmm"]
# A dictionary specifying which categories from the hists create the analysis categories
self.CATEGORIES = {
"cat_2lss_p": {
"channel": self.ch_2lss,
"sumcharge": ["ch+"],
"cut": ["1+bm2+bl"],
},
"cat_2lss_m": {
"channel": self.ch_2lss,
"sumcharge": ["ch-"],
"cut": ["1+bm2+bl"],
},
"cat_3l_1b_onZ": {
"channel": self.ch_3l_onZ,
"sumcharge": ["ch+", "ch-"],
"cut": ["1bm"],
},
"cat_3l_1b_offZ_p": {
"channel": self.ch_3l_offZ,
"sumcharge": ["ch+"],
"cut": ["1bm"],
},
"cat_3l_1b_offZ_m": {
"channel": self.ch_3l_offZ,
"sumcharge": ["ch-"],
"cut": ["1bm"],
},
"cat_3l_2b_onZ": {
"channel": self.ch_3l_onZ,
"sumcharge": ["ch+", "ch-"],
"cut": ["2+bm"],
},
"cat_3l_2b_offZ_p": {
"channel": self.ch_3l_offZ,
"sumcharge": ["ch+"],
"cut": ["2+bm"],
},
"cat_3l_2b_offZ_m": {
"channel": self.ch_3l_offZ,
"sumcharge": ["ch-"],
"cut": ["2+bm"],
},
"cat_4l": {
"channel": self.ch_4l,
"sumcharge": ["ch+", "ch-", "ch0"],
"cut": ["1+bm2+bl"],
},
}
# Yields from TOP-19-001 AN table 15
self.TOP19001_YLDS = {
"ttlnu": {
"cat_2lss_p": (81.1, None),
"cat_2lss_m": (44.0, None),
"cat_3l_1b_offZ_p": (16.6, None),
"cat_3l_1b_offZ_m": (9.1, None),
"cat_3l_2b_offZ_p": (12.1, None),
"cat_3l_2b_offZ_m": (6.7, None),
"cat_3l_1b_onZ": (3.4, None),
"cat_3l_2b_onZ": (2.5, None),
"cat_4l": (0.0, None),
},
"ttll": {
"cat_2lss_p": (22.6, None),
"cat_2lss_m": (22.5, None),
"cat_3l_1b_offZ_p": (14.2, None),
"cat_3l_1b_offZ_m": (14.7, None),
"cat_3l_2b_offZ_p": (10.1, None),
"cat_3l_2b_offZ_m": (9.4, None),
"cat_3l_1b_onZ": (106.5, None),
"cat_3l_2b_onZ": (70.9, None),
"cat_4l": (10.4, None),
},
"ttH": {
"cat_2lss_p": (28.6, None),
"cat_2lss_m": (27.9, None),
"cat_3l_1b_offZ_p": (8.5, None),
"cat_3l_1b_offZ_m": (8.1, None),
"cat_3l_2b_offZ_p": (5.5, None),
"cat_3l_2b_offZ_m": (5.6, None),
"cat_3l_1b_onZ": (3.5, None),
"cat_3l_2b_onZ": (2.4, None),
"cat_4l": (1.1, None),
},
"tllq": {
"cat_2lss_p": (2.9, None),
"cat_2lss_m": (1.7, None),
"cat_3l_1b_offZ_p": (3.8, None),
"cat_3l_1b_offZ_m": (1.9, None),
"cat_3l_2b_offZ_p": (1.3, None),
"cat_3l_2b_offZ_m": (0.6, None),
"cat_3l_1b_onZ": (42.1, None),
"cat_3l_2b_onZ": (14.1, None),
"cat_4l": (0.0, None),
},
"tHq": {
"cat_2lss_p": (0.9, None),
"cat_2lss_m": (0.5, None),
"cat_3l_1b_offZ_p": (0.3, None),
"cat_3l_1b_offZ_m": (0.2, None),
"cat_3l_2b_offZ_p": (0.2, None),
"cat_3l_2b_offZ_m": (0.1, None),
"cat_3l_1b_onZ": (0.1, None),
"cat_3l_2b_onZ": (0.1, None),
"cat_4l": (0.0, None),
},
}
'''
This script is used to transform scale factors, which are tipically provided as 2D histograms within root files,
into coffea format of corrections.
'''
#import uproot, uproot_methods
import uproot
from coffea import hist, lookup_tools
import os, sys
from topcoffea.modules.paths import topcoffea_path
def GetHistoFun(fname, hname):
f = uproot.open(fname)
h = f[hname]
return lookup_tools.dense_lookup.dense_lookup(h.values, h.edges)
getMuonIso = GetHistoFun(topcoffea_path('data/scaleFactors/MuonISO.root'),
'NUM_TightRelIso_DEN_TightIDandIPCut_pt_abseta')
getMuonId = GetHistoFun(topcoffea_path('data/scaleFactors/MuonID.root'),
'NUM_TightID_DEN_genTracks_pt_abseta')
Returns a dictionary containting all the info for each sample
Usage:
>> python samples.py configFile.cfg
Example of how to run this script in this repo:
>> python moca/samples.py cfg/2018.cfg
'''
import os, sys
from coffea.util import save
from topcoffea.modules.DASsearch import GetDatasetFromDAS
from topcoffea.modules.paths import topcoffea_path
from topcoffea.modules.fileReader import GetFiles, GetAllInfoFromFile
basepath = topcoffea_path("") # Just want path to topcoffea/topcoffea, not any particular file within it, so just pass "" to the function
def FindFileInDir(fname, dname = '.'):
if not os.path.isfile(dname+'/'+fname):
l = list(filter(lambda x: x[0] == fname, [x.split('.') for x in os.listdir(dname)]))
if len(l) == 0: return False
else : l = l[0]
fname = l[0] + '.' + l[1]
return fname
else: return dname+'/'+fname
def loadxsecdic(fname, verbose):
xsecdir = {}
dname = '.'
filename = FindFileInDir(fname, dname)
if not filename: filename = FindFileInDir(fname, basepath)
def get_lumi(year):
lumi_json = topcoffea_path("json/lumi.json")
with open(lumi_json) as f_lumi:
lumi = json.load(f_lumi)
lumi = lumi[year]
return lumi
import gzip
import pickle
from coffea.jetmet_tools import FactorizedJetCorrector, JetCorrectionUncertainty
from coffea.jetmet_tools import JECStack, CorrectedJetsFactory
from coffea.btag_tools.btagscalefactor import BTagScaleFactor
basepathFromTTH = 'data/fromTTH/lepSF/'
###### Lepton scale factors
################################################################
extLepSF = lookup_tools.extractor()
# Electron reco
extLepSF.add_weight_sets([
"ElecRecoSFb20_2016 EGamma_SF2D %s" %
topcoffea_path(basepathFromTTH +
'reco/elec/2016/el_scaleFactors_gsf_ptLt20.root')
])
extLepSF.add_weight_sets([
"ElecRecoSF_2016 EGamma_SF2D %s" %
topcoffea_path(basepathFromTTH +
'reco/elec/2016/el_scaleFactors_gsf_ptGt20.root')
])
extLepSF.add_weight_sets([
"ElecRecoSFb20_2017 EGamma_SF2D %s" %
topcoffea_path(basepathFromTTH +
'reco/elec/2017/el_scaleFactors_gsf_ptLt20.root')
])
extLepSF.add_weight_sets([
"ElecRecoSF_2017 EGamma_SF2D %s" %
topcoffea_path(basepathFromTTH +
'reco/elec/2017/el_scaleFactors_gsf_ptGt20.root')
import numpy as np import awkward as ak import gzip import pickle from coffea.jetmet_tools import FactorizedJetCorrector, JetCorrectionUncertainty from coffea.jetmet_tools import JECStack, CorrectedJetsFactory from coffea.btag_tools.btagscalefactor import BTagScaleFactor basepathFromTTH = 'data/fromTTH/lepSF/' ###### Lepton scale factors ################################################################ extLepSF = lookup_tools.extractor() # Electron reco extLepSF.add_weight_sets(["ElecRecoSFb20_2016 EGamma_SF2D %s"%topcoffea_path(basepathFromTTH+'reco/elec/2016/el_scaleFactors_gsf_ptLt20.root')]) extLepSF.add_weight_sets(["ElecRecoSF_2016 EGamma_SF2D %s"%topcoffea_path(basepathFromTTH+'reco/elec/2016/el_scaleFactors_gsf_ptGt20.root')]) extLepSF.add_weight_sets(["ElecRecoSFb20_2017 EGamma_SF2D %s"%topcoffea_path(basepathFromTTH+'reco/elec/2017/el_scaleFactors_gsf_ptLt20.root')]) extLepSF.add_weight_sets(["ElecRecoSF_2017 EGamma_SF2D %s"%topcoffea_path(basepathFromTTH+'reco/elec/2017/el_scaleFactors_gsf_ptGt20.root')]) extLepSF.add_weight_sets(["ElecRecoSF_2018 EGamma_SF2D %s"%topcoffea_path(basepathFromTTH+'reco/elec/2018/el_scaleFactors_gsf.root')]) extLepSF.add_weight_sets(["ElecRecoSFb20_2016_er EGamma_SF2D_error %s"%topcoffea_path(basepathFromTTH+'reco/elec/2016/el_scaleFactors_gsf_ptLt20.root')]) extLepSF.add_weight_sets(["ElecRecoSF_2016_er EGamma_SF2D_error %s"%topcoffea_path(basepathFromTTH+'reco/elec/2016/el_scaleFactors_gsf_ptGt20.root')]) extLepSF.add_weight_sets(["ElecRecoSFb20_2017_er EGamma_SF2D_error %s"%topcoffea_path(basepathFromTTH+'reco/elec/2017/el_scaleFactors_gsf_ptLt20.root')]) extLepSF.add_weight_sets(["ElecRecoSF_2017_er EGamma_SF2D_error %s"%topcoffea_path(basepathFromTTH+'reco/elec/2017/el_scaleFactors_gsf_ptGt20.root')]) extLepSF.add_weight_sets(["ElecRecoSF_2018_er EGamma_SF2D_error %s"%topcoffea_path(basepathFromTTH+'reco/elec/2018/el_scaleFactors_gsf.root')]) # Electron loose extLepSF.add_weight_sets(["ElecLooseSF_2016 EGamma_SF2D %s"%topcoffea_path(basepathFromTTH+'loose/elec/TnP_loose_ele_2016.root')]) extLepSF.add_weight_sets(["ElecLooseSF_2017 EGamma_SF2D %s"%topcoffea_path(basepathFromTTH+'loose/elec/TnP_loose_ele_2017.root')]) extLepSF.add_weight_sets(["ElecLooseSF_2018 EGamma_SF2D %s"%topcoffea_path(basepathFromTTH+'loose/elec/TnP_loose_ele_2018.root')]) extLepSF.add_weight_sets(["ElecLoosettHSF_2016 EGamma_SF2D %s"%topcoffea_path(basepathFromTTH+'loose/elec/TnP_loosettH_ele_2016.root')])