def histFitterAltSig(sample, tnpBin, tnpWorkspaceParam, isaddGaus=0):
tnpWorkspacePar = createWorkspaceForAltSig(sample, tnpBin,
tnpWorkspaceParam)
tnpWorkspaceFunc = [
"tailLeft[1]",
"RooCBExGaussShape::sigResPass(x,meanP,expr('sqrt(sigmaP*sigmaP+sosP*sosP)',{sigmaP,sosP}),alphaP,nP, expr('sqrt(sigmaP_2*sigmaP_2+sosP*sosP)',{sigmaP_2,sosP}),tailLeft)",
"RooCBExGaussShape::sigResFail(x,meanF,expr('sqrt(sigmaF*sigmaF+sosF*sosF)',{sigmaF,sosF}),alphaF,nF, expr('sqrt(sigmaF_2*sigmaF_2+sosF*sosF)',{sigmaF_2,sosF}),tailLeft)",
"RooCMSShape::bkgPass(x, acmsP, betaP, gammaP, peakP)",
"RooCMSShape::bkgFail(x, acmsF, betaF, gammaF, peakF)",
]
if isaddGaus == 1:
tnpWorkspaceFunc += [
"Gaussian::sigGaussFail(x,meanGF,sigmaGF)",
]
if sample.isMC:
tnpWorkspaceFunc += [
"sigFracF[0.5,0.0,1.0]",
]
tnpWorkspace = []
tnpWorkspace.extend(tnpWorkspacePar)
tnpWorkspace.extend(tnpWorkspaceFunc)
## init fitter
infile = rt.TFile(sample.histFile, "read")
hP = infile.Get('%s_Pass' % tnpBin['name'])
hF = infile.Get('%s_Fail' % tnpBin['name'])
## for high pT change the failing spectra to passing probe to get statistics
## MC only: this is to get MC parameters in data fit!
if sample.isMC and ptMin(tnpBin) > minPtForSwitch:
hF = infile.Get('%s_Pass' % tnpBin['name'])
fitter = tnpFitter(hP, hF, tnpBin['name'])
# fitter.fixSigmaFtoSigmaP()
infile.Close()
## setup
rootfile = rt.TFile(sample.altSigFit, 'update')
fitter.setOutputFile(rootfile)
## generated Z LineShape
fileTruth = rt.TFile('etc/inputs/ZeeGenLevel.root', 'read')
histZLineShape = fileTruth.Get('Mass')
fitter.setZLineShapes(histZLineShape, histZLineShape)
fileTruth.Close()
### set workspace
workspace = rt.vector("string")()
for iw in tnpWorkspace:
workspace.push_back(iw)
fitter.setWorkspace(workspace, isaddGaus)
title = tnpBin['title'].replace(';', ' - ')
title = title.replace('probe_sc_eta', '#eta_{SC}')
title = title.replace('probe_Ele_pt', 'p_{T}')
fitter.fits(sample.mcTruth, title, isaddGaus)
rootfile.Close()
def histFitterAltBkg(sample, tnpBin, tnpWorkspaceParam):
tnpWorkspaceFunc = [
"Gaussian::sigResPass(x,meanP,sigmaP)",
"Gaussian::sigResFail(x,meanF,sigmaF)",
"Exponential::bkgPass(x, alphaP)",
"Exponential::bkgFail(x, alphaF)",
# "RooPowerLaw::bkgPass(x, acmsP, betaP, gammaP, peakP)",#bing
# "RooPowerLaw::bkgFail(x, acmsF, betaF, gammaF, peakF)",#bing
# "RooDoubleExp::bkgFail(x, acmsF, betaF, gammaF, peakF)",#bing
# "RooTripleExp::bkgPass(x, acmsP, betaP, gammaP, peakP, p1P, p2P)",#bing
# "RooTripleExp::bkgFail(x, acmsF, betaF, gammaF, peakF, p1F, p2F)",#bing
# "RooCMSShapeExp::bkgPass(x, acmsP, betaP, gammaP, peakP, p1P, p2P)",#bing
# "RooCMSShapeExp::bkgFail(x, acmsF, betaF, gammaF, peakF, p1F, p2F)",#bing
]
tnpWorkspace = []
tnpWorkspace.extend(tnpWorkspaceParam)
tnpWorkspace.extend(tnpWorkspaceFunc)
# init fitter
infile = rt.TFile(sample.histFile, 'read')
hP = infile.Get('%s_Pass' % tnpBin['name'])
hF = infile.Get('%s_Fail' % tnpBin['name'])
fitter = tnpFitter(hP, hF, tnpBin['name'])
infile.Close()
# setup
rootfile = rt.TFile(sample.altBkgFit, 'update')
fitter.setOutputFile(rootfile)
# fitter.setFitRange(65,115)
# generated Z LineShape
# for high pT change the failing spectra to any probe to get statistics
fileTruth = rt.TFile(sample.mcRef.histFile, 'read')
histZLineShapeP = fileTruth.Get('%s_Pass' % tnpBin['name'])
histZLineShapeF = fileTruth.Get('%s_Fail' % tnpBin['name'])
if ptMin(tnpBin) > minPtForSwitch:
histZLineShapeF = fileTruth.Get('%s_Pass' % tnpBin['name'])
# fitter.fixSigmaFtoSigmaP()
fitter.setZLineShapes(histZLineShapeP, histZLineShapeF)
fileTruth.Close()
# set workspace
workspace = rt.vector("string")()
for iw in tnpWorkspace:
workspace.push_back(iw)
fitter.setWorkspace(workspace)
title = tnpBin['title'].replace(';', ' - ')
title = title.replace('probe_sc_eta', '#eta_{SC}')
title = title.replace('probe_Ele_pt', 'p_{T}')
fitter.fits(sample.mcTruth, title)
rootfile.Close()
def histFitterAltSig( sample, tnpBin, tnpWorkspaceParam ):
tnpWorkspacePar = createWorkspaceForAltSig( sample, tnpBin, tnpWorkspaceParam )
tnpWorkspaceFunc = [
"tailLeft[1]",
"RooCBExGaussShape::sigResPass(x,meanP,expr('sqrt(sigmaP*sigmaP+sosP*sosP)',{sigmaP,sosP}),alphaP,nP, expr('sqrt(sigmaP_2*sigmaP_2+sosP*sosP)',{sigmaP_2,sosP}),tailLeft)",
"RooCBExGaussShape::sigResFail(x,meanF,expr('sqrt(sigmaF*sigmaF+sosF*sosF)',{sigmaF,sosF}),alphaF,nF, expr('sqrt(sigmaF_2*sigmaF_2+sosF*sosF)',{sigmaF_2,sosF}),tailLeft)",
"RooCMSShape::bkgPass(x, acmsP, betaP, gammaP, peakP)",
"RooCMSShape::bkgFail(x, acmsF, betaF, gammaF, peakF)",
]
tnpWorkspace = []
tnpWorkspace.extend(tnpWorkspacePar)
tnpWorkspace.extend(tnpWorkspaceFunc)
## init fitter
infile = rt.TFile( sample.histFile, "read")
hP = infile.Get('%s_Pass' % tnpBin['name'] )
hF = infile.Get('%s_Fail' % tnpBin['name'] )
## for high pT change the failing spectra to passing probe to get statistics
## MC only: this is to get MC parameters in data fit!
if sample.isMC and ptMin( tnpBin ) > minPtForSwitch:
hF = infile.Get('%s_Pass' % tnpBin['name'] )
fitter = tnpFitter( hP, hF, tnpBin['name'] )
# fitter.fixSigmaFtoSigmaP()
infile.Close()
## setup
rootfile = rt.TFile(sample.altSigFit,'update')
fitter.setOutputFile( rootfile )
## generated Z LineShape
fileTruth = rt.TFile('etc/inputs/ZeeGenLevel.root','read')
histZLineShape = fileTruth.Get('Mass')
fitter.setZLineShapes(histZLineShape,histZLineShape)
fileTruth.Close()
### set workspace
workspace = rt.vector("string")()
for iw in tnpWorkspace:
workspace.push_back(iw)
fitter.setWorkspace( workspace )
title = tnpBin['title'].replace(';',' - ')
title = title.replace('probe_sc_eta','#eta_{SC}')
title = title.replace('probe_Ele_pt','p_{T}')
fitter.fits(sample.mcTruth,title)
rootfile.Close()
def histFitterNominal( sample, tnpBin, tnpWorkspaceParam ):
tnpWorkspaceFunc = [
"Gaussian::sigResPass(x,meanP,sigmaP)",
"Gaussian::sigResFail(x,meanF,sigmaF)",
"RooCMSShape::bkgPass(x, acmsP, betaP, gammaP, peakP)",
"RooCMSShape::bkgFail(x, acmsF, betaF, gammaF, peakF)",
]
tnpWorkspace = []
tnpWorkspace.extend(tnpWorkspaceParam)
tnpWorkspace.extend(tnpWorkspaceFunc)
## init fitter
infile = rt.TFile( sample.histFile, "read")
hP = infile.Get('%s_Pass' % tnpBin['name'] )
hF = infile.Get('%s_Fail' % tnpBin['name'] )
fitter = tnpFitter( hP, hF, tnpBin['name'] )
infile.Close()
## setup
fitter.useMinos()
rootfile = rt.TFile(sample.nominalFit,'update')
fitter.setOutputFile( rootfile )
## generated Z LineShape
## for high pT change the failing spectra to any probe to get statistics
fileTruth = rt.TFile(sample.mcRef.histFile,'read')
histZLineShapeP = fileTruth.Get('%s_Pass'%tnpBin['name'])
histZLineShapeF = fileTruth.Get('%s_Fail'%tnpBin['name'])
if ptMin( tnpBin ) > minPtForSwitch:
histZLineShapeF = fileTruth.Get('%s_Pass'%tnpBin['name'])
# fitter.fixSigmaFtoSigmaP()
fitter.setZLineShapes(histZLineShapeP,histZLineShapeF)
fileTruth.Close()
### set workspace
workspace = rt.vector("string")()
for iw in tnpWorkspace:
workspace.push_back(iw)
fitter.setWorkspace( workspace )
title = tnpBin['title'].replace(';',' - ')
title = title.replace('probe_sc_eta','#eta_{SC}')
title = title.replace('probe_Ele_pt','p_{T}')
fitter.fits(sample.mcTruth,title)
rootfile.Close()
def histFitterAltSig( sample, tnpBin, tnpWorkspaceParam ):
tnpWorkspacePar = createWorkspaceForAltSig( sample, tnpBin, tnpWorkspaceParam )
tnpWorkspaceFunc = [
"RooCBExGaussShape::sigResPass(x,meanP,expr('sqrt(sigmaP*sigmaP+sosP*sosP)',{sigmaP,sosP}),alphaP,nP, expr('sqrt(sigmaP_2*sigmaP_2+sosP*sosP)',{sigmaP_2,sosP}))",
"RooCBExGaussShape::sigResFail(x,meanF,expr('sqrt(sigmaF*sigmaF+sosF*sosF)',{sigmaF,sosF}),alphaF,nF, expr('sqrt(sigmaF_2*sigmaF_2+sosF*sosF)',{sigmaF_2,sosF}))",
"RooCMSShape::bkgPass(x, acmsP, betaP, gammaP, peakP)",
"RooCMSShape::bkgFail(x, acmsF, betaF, gammaF, peakF)",
]
tnpWorkspace = []
tnpWorkspace.extend(tnpWorkspacePar)
tnpWorkspace.extend(tnpWorkspaceFunc)
## init fitter
infile = rt.TFile(sample.histFile,'read')
fitter = tnpFitter(infile, tnpBin['name'] )
infile.Close()
rootfile = rt.TFile(sample.altSigFit,'update')
fitter.setOutputFile( rootfile )
# ptmin = ptMin(tnpBin)
# fitter.setMin(2*math.sqrt(ptmin*30) )
## generated Z LineShape
fileTruth = rt.TFile('etc/inputs/ZeeGenLevel.root','read')
histZLineShape = fileTruth.Get('Mass')
fitter.setZLineShapes(histZLineShape,histZLineShape)
fileTruth.Close()
### set workspace
workspace = rt.vector("string")()
for iw in tnpWorkspace:
workspace.push_back(iw)
fitter.setWorkspace( workspace )
title = tnpBin['title'].replace(';',' - ')
title = title.replace('probe_sc_eta','#eta_{SC}')
title = title.replace('probe_Ele_pt','p_{T}')
fitter.fits(sample.mcTruth,title)
rootfile.Close()
def histFitterNominal( sample, tnpBin, tnpWorkspaceParam ):
tnpWorkspaceFunc = [
"Gaussian::sigResPass(x,meanP,sigmaP)",
"Gaussian::sigResFail(x,meanF,sigmaF)",
"RooCMSShape::bkgPass(x, acmsP, betaP, gammaP, peakP)",
"RooCMSShape::bkgFail(x, acmsF, betaF, gammaF, peakF)",
]
tnpWorkspace = []
tnpWorkspace.extend(tnpWorkspaceParam)
tnpWorkspace.extend(tnpWorkspaceFunc)
## init fitter
infile = rt.TFile( sample.histFile, "read")
fitter = tnpFitter( infile, tnpBin['name'] )
fitter.useMinos()
infile.Close()
rootfile = rt.TFile(sample.nominalFit,'update')
fitter.setOutputFile( rootfile )
## generated Z LineShape
fileTruth = rt.TFile(sample.mcRef.histFile,'read')
histZLineShapeP = fileTruth.Get('%s_Pass'%tnpBin['name'])
histZLineShapeF = fileTruth.Get('%s_Fail'%tnpBin['name'])
fitter.setZLineShapes(histZLineShapeP,histZLineShapeF)
fileTruth.Close()
### set workspace
workspace = rt.vector("string")()
for iw in tnpWorkspace:
workspace.push_back(iw)
fitter.setWorkspace( workspace )
title = tnpBin['title'].replace(';',' - ')
title = title.replace('probe_sc_eta','#eta_{SC}')
title = title.replace('probe_Ele_pt','p_{T}')
fitter.fits(sample.mcTruth,title)
rootfile.Close()