def plotMPVs(self,fitList):
x = np.arange(-2,3)
y = np.arange(-2,3)
z = []
for fit in fitList:
z.append(fit.GetParameter(1))
z = np.reshape(z, [len(x), len(y)])
z = np.flipud(z)
fig, ax = plt.subplots()
im = ax.pcolor(z)
ax.set_xticks(np.arange(z.shape[0])+0.5, minor=False)
ax.set_yticks(np.arange(z.shape[1])+0.5, minor=False)
ax.set_xticklabels(x, minor=False)
ax.set_yticklabels(y, minor=False)
colorbar = fig.colorbar(im)
colorbar.set_label('MPV reconstructed energy / GeV')
ax.axis('tight')
pyplotCmsPrivateLabel(ax,y=1)
plt.xlabel(r'$\Delta i\eta$')
plt.ylabel(r'$\Delta i\phi$')
plt.savefig('plots/averageEnergy/mpv.png')
plt.show()
def plotDeltaNL1ComparedMatchingByEMax(self):
effL1Muon = self.fileHandler.getHistogram("hoMuonAnalyzer/efficiency/GenAndL1MuonPt15_Efficiency")
effL1MuonTruth = self.fileHandler.getHistogram("hoMuonAnalyzer/efficiency/GenAndL1MuonAndHoAboveThrPt15_Efficiency")
xValues = []
xErrLow = []
xErrHigh= []
yValues = []
for i in range(effL1Muon.GetPassedHistogram().GetNbinsX()):
if(effL1Muon.GetTotalHistogram().GetBinContent(i) != 0):
yValues.append(effL1MuonTruth.GetTotalHistogram().GetBinContent(i) - effL1Muon.GetTotalHistogram().GetBinContent(i))
xValues.append(effL1Muon.GetTotalHistogram().GetBinCenter(i))
xErrLow.append(0.25)
xErrHigh.append(.25)
for i in range(1,len(xValues) -1 ):
xErrLow.append((xValues[i] - xValues[i-1])/2.)
xErrHigh.append((xValues[i+1] - xValues[i])/2.)
xErrLow.append(xErrLow[-1])
xErrHigh.append(xErrHigh[-1])
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax1.errorbar(xValues,yValues,xerr=[xErrLow,xErrHigh],fmt='^',color='#00549F',label='(L1 + HO) - L1')
ax1.legend(loc='lower right')
ax1.grid(True)
ax1.set_ylabel('# L1')
ax1.set_xlabel(r'p$_\mathrm{T}$ / GeV')
ax1.set_title(r'# L1 in direct comparison for $\Delta$R matching by E$_{Max}$')
pyplotCmsPrivateLabel(ax1)
plt.savefig('plots/efficiency/nL1ByEmax.png')
plt.show()
def plotMPVs(self,fitList):
x = np.arange(-2,3)
y = np.arange(-2,3)
z = []
for fit in fitList:
z.append(fit.GetParameter(1))
z = np.reshape(z, [len(x), len(y)])
z = np.flipud(z)
fig, ax = plt.subplots()
im = ax.pcolor(z)
ax.set_xticks(np.arange(z.shape[0])+0.5, minor=False)
ax.set_yticks(np.arange(z.shape[1])+0.5, minor=False)
ax.set_xticklabels(x, minor=False)
ax.set_yticklabels(y, minor=False)
colorbar = fig.colorbar(im)
colorbar.set_label('MPV reconstructed energy / GeV')
ax.axis('tight')
pyplotCmsPrivateLabel(ax,y=1)
plt.xlabel(r'$\Delta i\eta$')
plt.ylabel(r'$\Delta i\phi$')
plt.savefig('plots/averageEnergy/mpv.png')
plt.show()
def plotDeltaNL1ComparedGridMatching(self):
effL1Muon3x3Truth = self.fileHandler.getHistogram('hoMuonAnalyzer/efficiency/L1MuonTruth3x3_Efficiency')
effL1Muon3x3 = self.fileHandler.getHistogram('hoMuonAnalyzer/efficiency/L1Muon3x3_Efficiency')
xValues = []
xErrLow = []
xErrHigh= []
yValues = []
for i in range(effL1Muon3x3.GetPassedHistogram().GetNbinsX()):
if(effL1Muon3x3.GetTotalHistogram().GetBinContent(i) != 0):
yValues.append(effL1Muon3x3Truth.GetTotalHistogram().GetBinContent(i) - effL1Muon3x3.GetTotalHistogram().GetBinContent(i))
xValues.append(effL1Muon3x3.GetTotalHistogram().GetBinCenter(i))
xErrLow.append(0.25)
xErrHigh.append(.25)
for i in range(1,len(xValues) -1 ):
xErrLow.append((xValues[i] - xValues[i-1])/2.)
xErrHigh.append((xValues[i+1] - xValues[i])/2.)
xErrLow.append(xErrLow[-1])
xErrHigh.append(xErrHigh[-1])
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax1.errorbar(xValues,yValues,xerr=[xErrLow,xErrHigh],fmt='^',color='#57AB27',label='# Truth - # All')
ax1.legend(loc='lower right')
ax1.grid(True)
ax1.set_ylabel('# L1')
ax1.set_xlabel(r'p$_\mathrm{T}$ / GeV')
ax1.set_title(r'# L1 in direct comparison for Grid Matching (3x3)')
pyplotCmsPrivateLabel(ax1)
plt.savefig('plots/efficiency/nL13x3.png')
plt.show()
def plotDeltaNL1ComparedGridMatching(self):
effL1Muon3x3Truth = self.fileHandler.getHistogram(
'efficiency/L1MuonTruth3x3_Efficiency')
effL1Muon3x3 = self.fileHandler.getHistogram(
'efficiency/L1Muon3x3_Efficiency')
xValues = []
xErrLow = []
xErrHigh = []
yValues = []
for i in range(effL1Muon3x3.GetPassedHistogram().GetNbinsX()):
if (effL1Muon3x3.GetTotalHistogram().GetBinContent(i) != 0):
yValues.append(
effL1Muon3x3Truth.GetTotalHistogram().GetBinContent(i) -
effL1Muon3x3.GetTotalHistogram().GetBinContent(i))
xValues.append(
effL1Muon3x3.GetTotalHistogram().GetBinCenter(i))
xErrLow.append(0.25)
xErrHigh.append(.25)
for i in range(1, len(xValues) - 1):
xErrLow.append((xValues[i] - xValues[i - 1]) / 2.)
xErrHigh.append((xValues[i + 1] - xValues[i]) / 2.)
xErrLow.append(xErrLow[-1])
xErrHigh.append(xErrHigh[-1])
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax1.errorbar(xValues,
yValues,
xerr=[xErrLow, xErrHigh],
fmt='^',
color='#57AB27',
label='# Truth - # All')
ax1.legend(loc='lower right')
ax1.grid(True)
ax1.set_ylabel('# L1')
ax1.set_xlabel(r'p$_\mathrm{T}$ / GeV')
ax1.set_title(r'# L1 in direct comparison for Grid Matching (3x3)')
pyplotCmsPrivateLabel(ax1)
plt.savefig('plots/nL13x3.png')
plt.show(block=False)
def plotDeltaNL1ComparedMatchingByEMax(self):
effL1Muon = self.fileHandler.getHistogram(
"efficiency/GenAndL1MuonPt15_Efficiency")
effL1MuonTruth = self.fileHandler.getHistogram(
"efficiency/GenAndL1MuonAndHoAboveThrPt15_Efficiency")
xValues = []
xErrLow = []
xErrHigh = []
yValues = []
for i in range(effL1Muon.GetPassedHistogram().GetNbinsX()):
if (effL1Muon.GetTotalHistogram().GetBinContent(i) != 0):
yValues.append(
effL1MuonTruth.GetTotalHistogram().GetBinContent(i) -
effL1Muon.GetTotalHistogram().GetBinContent(i))
xValues.append(effL1Muon.GetTotalHistogram().GetBinCenter(i))
xErrLow.append(0.25)
xErrHigh.append(.25)
for i in range(1, len(xValues) - 1):
xErrLow.append((xValues[i] - xValues[i - 1]) / 2.)
xErrHigh.append((xValues[i + 1] - xValues[i]) / 2.)
xErrLow.append(xErrLow[-1])
xErrHigh.append(xErrHigh[-1])
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax1.errorbar(xValues,
yValues,
xerr=[xErrLow, xErrHigh],
fmt='^',
color='#00549F',
label='(L1 + HO) - L1')
ax1.legend(loc='lower right')
ax1.grid(True)
ax1.set_ylabel('# L1')
ax1.set_xlabel(r'p$_\mathrm{T}$ / GeV')
ax1.set_title(
r'# L1 in direct comparison for $\Delta$R matching by E$_{Max}$')
pyplotCmsPrivateLabel(ax1)
plt.savefig('plots/nL1ByEmax.png')
plt.show(block=False)
def plotL1GridMatchingEfficiency(self):
effL1MuonCentral = self.fileHandler.getHistogram('hoMuonAnalyzer/efficiency/L1MuonCentral_Efficiency')
effL1Muon3x3 = self.fileHandler.getHistogram('hoMuonAnalyzer/efficiency/L1Muon3x3_Efficiency')
effL1Muon5x5 = self.fileHandler.getHistogram('hoMuonAnalyzer/efficiency/L1Muon5x5_Efficiency')
c = TCanvas("c1","c1",1200,1200)
effL1Muon5x5.SetMarkerColor(colorRwthMagenta)
effL1Muon5x5.SetLineColor(colorRwthMagenta)
effL1Muon5x5.SetMarkerStyle(27)
effL1Muon5x5.SetTitle('Efficiency for Matching HO to L1;p_{T} / GeV;Efficiency')
effL1Muon5x5.Draw('')
c.Update()
effL1Muon5x5.GetPaintedGraph().GetYaxis().SetRangeUser(0.4,1)
effL1Muon3x3.SetMarkerColor(colorRwthGruen)
effL1Muon3x3.SetLineColor(colorRwthGruen)
effL1Muon3x3.SetMarkerStyle(26)
effL1Muon3x3.Draw("same")
effL1MuonCentral.SetMarkerColor(colorRwthDarkBlue)
effL1MuonCentral.SetLineColor(colorRwthDarkBlue)
effL1MuonCentral.SetMarkerStyle(4)
effL1MuonCentral.Draw("same")
c.Update()
setupAxes(effL1MuonCentral)
setupAxes(effL1Muon3x3)
setupAxes(effL1Muon5x5)
legend = TLegend(0.55,0.1,0.9,0.3)
legend.AddEntry(effL1MuonCentral,'Matches in central tile','ep')
legend.AddEntry(effL1Muon3x3,'Matches in 3x3 grid','ep')
legend.AddEntry(effL1Muon5x5,'Matches in 5x5 grid','ep')
legend.Draw()
effCentral = []
eff3x3 = []
eff5x5 = []
xValues = []
xErrLow = []
xErrHigh = []
yErrCentralHigh = []
yErrCentralLow = []
yErr3x3High = []
yErr3x3Low = []
yErr5x5High = []
yErr5x5Low = []
for i in range(effL1MuonCentral.GetPassedHistogram().GetNbinsX()):
if effL1MuonCentral.GetTotalHistogram().GetBinContent(i) != 0:
effCentral.append(effL1MuonCentral.GetPassedHistogram().GetBinContent(i)/effL1MuonCentral.GetTotalHistogram().GetBinContent(i)*100)
xValues.append(effL1MuonCentral.GetTotalHistogram().GetBinCenter(i))
if effL1Muon3x3.GetTotalHistogram().GetBinContent(i) != 0:
eff3x3.append(effL1Muon3x3.GetPassedHistogram().GetBinContent(i)/effL1Muon3x3.GetTotalHistogram().GetBinContent(i)*100)
if effL1Muon5x5.GetTotalHistogram().GetBinContent(i) != 0:
eff5x5.append(effL1Muon5x5.GetPassedHistogram().GetBinContent(i)/effL1Muon5x5.GetTotalHistogram().GetBinContent(i)*100)
for i in range(effL1MuonCentral.GetPaintedGraph().GetN()):
yErrCentralHigh.append(effL1MuonCentral.GetPaintedGraph().GetErrorYhigh(i)*100)
yErrCentralLow.append(effL1MuonCentral.GetPaintedGraph().GetErrorYlow(i)*100)
yErr3x3High.append(effL1Muon3x3.GetPaintedGraph().GetErrorYhigh(i)*100)
yErr3x3Low.append(effL1Muon3x3.GetPaintedGraph().GetErrorYlow(i)*100)
yErr5x5High.append(effL1Muon5x5.GetPaintedGraph().GetErrorYhigh(i)*100)
yErr5x5Low.append(effL1Muon5x5.GetPaintedGraph().GetErrorYlow(i)*100)
xErrLow.append(0.25)
xErrHigh.append(.25)
for i in range(1,len(xValues) -1 ):
xErrLow.append((xValues[i] - xValues[i-1])/2.)
xErrHigh.append((xValues[i+1] - xValues[i])/2.)
xErrLow.append(xErrLow[-1])
xErrHigh.append(xErrHigh[-1])
fig = plt.figure()
ax = fig.add_subplot(1,1,1)
plt.errorbar(xValues,effCentral,yerr = [yErrCentralLow,yErrCentralHigh] ,xerr=[xErrLow,xErrHigh],fmt='s',color='#00549F',label='Match in central tile')
plt.errorbar(xValues,eff3x3,yerr = [yErr3x3Low,yErr3x3High] ,xerr=[xErrLow,xErrHigh],fmt='o',color='#E30066',label='Match in 3x3')
plt.errorbar(xValues,eff5x5,yerr = [yErr5x5Low,yErr5x5High] ,xerr=[xErrLow,xErrHigh],fmt='^',color='#57AB27',label='Match in 5x5')
plt.gca().xaxis.grid(True)
plt.gca().yaxis.grid(True)
plt.axis([0.,155,40,101])
plt.xlabel(r'p$_{\mathrm{T}}$ / GeV')
plt.ylabel(r'$\epsilon$ / %')
plt.title(r'Efficiency $\epsilon$ of matching L1 to HO')
plt.legend(loc='lower right')
pyplotCmsPrivateLabel(ax)
plt.savefig('plots/efficiency/efficiency.png')
plt.show()
def plot3x3GridTogether(self):
c = TCanvas()
effL1Muon3x3Truth = self.fileHandler.getHistogram('hoMuonAnalyzer/efficiency/L1MuonTruth3x3_Efficiency')
effL1Muon3x3 = self.fileHandler.getHistogram('hoMuonAnalyzer/efficiency/L1Muon3x3_Efficiency')
effL1Muon3x3.Draw()
effL1Muon3x3Truth.Draw("same")
c.Update()
xValues = []
xErrLow = []
xErrHigh = []
eff5x5 = []
yErr3x3High = []
yErr3x3Low = []
effTruth3x3 = []
yErr3x3TruthHigh = []
yErr3x3TruthLow = []
diff = []
diffErrHigh = []
diffErrLow = []
for i in range(effL1Muon3x3.GetPassedHistogram().GetNbinsX()):
if effL1Muon3x3.GetTotalHistogram().GetBinContent(i) != 0:
eff5x5.append(effL1Muon3x3.GetPassedHistogram().GetBinContent(i)/effL1Muon3x3.GetTotalHistogram().GetBinContent(i)*100)
xValues.append(effL1Muon3x3.GetTotalHistogram().GetBinCenter(i))
if effL1Muon3x3Truth.GetTotalHistogram().GetBinContent(i) != 0:
effTruth3x3.append(effL1Muon3x3Truth.GetPassedHistogram().GetBinContent(i)/effL1Muon3x3Truth.GetTotalHistogram().GetBinContent(i)*100)
for i in range(effL1Muon3x3.GetPaintedGraph().GetN()):
yErr3x3High.append(effL1Muon3x3.GetPaintedGraph().GetErrorYhigh(i)*100)
yErr3x3Low.append(effL1Muon3x3.GetPaintedGraph().GetErrorYlow(i)*100)
yErr3x3TruthHigh.append(effL1Muon3x3Truth.GetPaintedGraph().GetErrorYhigh(i)*100)
yErr3x3TruthLow.append(effL1Muon3x3Truth.GetPaintedGraph().GetErrorYlow(i)*100)
xErrLow.append(0.25)
xErrHigh.append(.25)
for i in range(0,len(eff5x5)):
diff.append(effTruth3x3[i] - eff5x5[i])
diffErrHigh.append(sqrt(pow(yErr3x3High[i], 2) + pow(yErr3x3TruthHigh[i], 2)))
diffErrLow.append(sqrt(pow(yErr3x3Low[i], 2) + pow(yErr3x3TruthLow[i], 2)))
for i in range(1,len(xValues) -1 ):
xErrLow.append((xValues[i] - xValues[i-1])/2.)
xErrHigh.append((xValues[i+1] - xValues[i])/2.)
xErrLow.append(xErrLow[-1])
xErrHigh.append(xErrHigh[-1])
fig = plt.figure()
ax1 = fig.add_subplot(211)
ax2 = fig.add_subplot(212,sharex=ax1)
ax1.errorbar(xValues,eff5x5,yerr = [yErr3x3Low,yErr3x3High] ,xerr=[xErrLow,xErrHigh],fmt='^',color='#57AB27',label='L1')
ax1.errorbar(xValues,effTruth3x3,yerr = [yErr3x3TruthLow,yErr3x3TruthHigh] ,xerr=[xErrLow,xErrHigh],fmt='^',color='#E30066',label='Truth L1')
ax1.legend(loc='lower right')
ax1.grid(True)
ax1.set_ylabel(r'$\epsilon$ / %')
ax1.set_title(r'Efficiency $\epsilon$ for 3x3 grid in direct comparison')
pyplotCmsPrivateLabel(ax1)
ax2.errorbar(xValues,diff,yerr=[diffErrLow,diffErrHigh], xerr=[xErrLow,xErrHigh],fmt='o', color='#00549F', label=r'$\epsilon$(Truth L1) - $\epsilon$(L1)')
ax2.legend()
ax2.grid(True)
ax2.set_ylabel(r'$\Delta$ $\epsilon$ / %')
ax2.set_xlabel(r'p$_\mathrm{T}$ / GeV')
#make x labels of first plot invisible
plt.setp(ax1.get_xticklabels(),visible=False)
#avoid overlapping labels on y axis
plt.setp(ax1.get_yticklabels()[0],visible=False)
plt.setp(ax2.get_yticklabels()[-1],visible=False)
fig.subplots_adjust(hspace=0)
plt.savefig('plots/efficiency/efficiency3x3.png')
plt.show()
def plotNtotalGridMatching3x3(self):
effL1Muon3x3 = self.fileHandler.getHistogram("hoMuonAnalyzer/efficiency/L1Muon3x3_Efficiency")
effL1Muon3x3Truth = self.fileHandler.getHistogram("hoMuonAnalyzer/efficiency/L1MuonTruth3x3_Efficiency")
effL1Muon5x5Truth = self.fileHandler.getHistogram("hoMuonAnalyzer/efficiency/L1MuonTruth5x5_Efficiency")
genPt = self.fileHandler.getHistogram("hoMuonAnalyzer/gen_Pt")
yTruth = []
yTruthErr = []
xValues = []
xErrLow = []
xErrHigh= []
yValues = []
xGen = []
xGenErr = []
yGen = []
yGenSummed = []
yGenSummedErr = []
for i in range(1,genPt.GetNbinsX()):
if(genPt.GetBinCenter(i) > 200):
break
xGen.append(genPt.GetBinCenter(i))
xGenErr.append(genPt.GetBinWidth(i)/2.)
yGen.append( genPt.GetBinContent(i) )
for i in range(1,effL1Muon3x3.GetPassedHistogram().GetNbinsX()):
if(effL1Muon3x3.GetTotalHistogram().GetBinContent(i) != 0):
yValues.append(effL1Muon3x3.GetTotalHistogram().GetBinContent(i))
xValues.append(effL1Muon3x3.GetTotalHistogram().GetBinCenter(i))
yTruth.append(effL1Muon3x3Truth.GetTotalHistogram().GetBinContent(i))
yTruthErr.append(sqrt(yTruth[-1]))
print effL1Muon3x3.GetTotalHistogram().Integral(),effL1Muon3x3Truth.GetTotalHistogram().Integral()\
,genPt.Integral(),effL1Muon5x5Truth.GetTotalHistogram().Integral()
xErrLow.append(0.25)
xErrHigh.append(.25)
for i in range(1,len(xValues) -1 ):
xErrLow.append((xValues[i] - xValues[i-1])/2.)
xErrHigh.append((xValues[i+1] - xValues[i])/2.)
xErrLow.append(xErrLow[-1])
xErrHigh.append(60)#xErrHigh[-1])
counter = 0
genSum = 0
for i in range(len(xGen)):
if(xGen[i]>(xValues[counter] + xErrHigh[counter])):
counter+=1
yGenSummed.append(genSum)
yGenSummedErr.append(sqrt(yGenSummed[-1]))
genSum = 0
genSum += yGen[i]
yGenSummed.append(genSum)
yGenSummedErr.append(sqrt(yGenSummed[-1]))
for i in range(0,len(yValues)):
yValues[i] /= (xErrHigh[i] + xErrLow[i])
for i in range(0,len(yTruth)):
yTruth[i] /= (xErrHigh[i] + xErrLow[i])
yTruthErr[i] /= (xErrHigh[i] + xErrLow[i])
for i in range(0,len(yGenSummed)):
yGenSummed[i] /= (xErrHigh[i] + xErrLow[i])
yGenSummedErr[i] /= (xErrHigh[i] + xErrLow[i])
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax1.errorbar(xValues,yValues,xerr=[xErrLow,xErrHigh],fmt='^',color=rwthDarkBlue,label='L1')
ax1.set_yscale('log')
ax1.errorbar(xValues,yTruth,xerr=[xErrLow,xErrHigh],yerr=yTruthErr,fmt='^',color=rwthGruen,label='L1 Truth')
# ax1.errorbar(xGen,yGen,xerr=xGenErr,fmt='^',color=rwthMagenta,label='Gen')
ax1.errorbar(xValues,yGenSummed,xerr=[xErrLow,xErrHigh],yerr = yGenSummedErr,fmt='^',color=rwthRot,label='Gen Summed')
ax1.legend(loc='lower right')
ax1.grid(True)
ax1.set_ylabel('# / bin width (1/GeV)')
ax1.set_xlabel(r'p$_\mathrm{T}$ / GeV')
ax1.set_title('# L1')
pyplotCmsPrivateLabel(ax1)
plt.savefig('plots/efficiency/nL1ByEmax3x3Absolute.png')
plt.show()
def plot3x3GridTogether(self):
c = TCanvas()
effL1Muon3x3Truth = self.fileHandler.getHistogram(
'efficiency/L1MuonTruth3x3_Efficiency')
effL1Muon3x3 = self.fileHandler.getHistogram(
'efficiency/L1Muon3x3_Efficiency')
effL1Muon3x3.Draw()
effL1Muon3x3Truth.Draw("same")
c.Update()
xValues = []
xErrLow = []
xErrHigh = []
eff5x5 = []
yErr3x3High = []
yErr3x3Low = []
effTruth3x3 = []
yErr3x3TruthHigh = []
yErr3x3TruthLow = []
diff = []
diffErrHigh = []
diffErrLow = []
for i in range(effL1Muon3x3.GetPassedHistogram().GetNbinsX()):
if effL1Muon3x3.GetTotalHistogram().GetBinContent(i) != 0:
eff5x5.append(
effL1Muon3x3.GetPassedHistogram().GetBinContent(i) /
effL1Muon3x3.GetTotalHistogram().GetBinContent(i) * 100)
xValues.append(
effL1Muon3x3.GetTotalHistogram().GetBinCenter(i))
if effL1Muon3x3Truth.GetTotalHistogram().GetBinContent(i) != 0:
effTruth3x3.append(
effL1Muon3x3Truth.GetPassedHistogram().GetBinContent(i) /
effL1Muon3x3Truth.GetTotalHistogram().GetBinContent(i) *
100)
for i in range(effL1Muon3x3.GetPaintedGraph().GetN()):
yErr3x3High.append(
effL1Muon3x3.GetPaintedGraph().GetErrorYhigh(i) * 100)
yErr3x3Low.append(effL1Muon3x3.GetPaintedGraph().GetErrorYlow(i) *
100)
yErr3x3TruthHigh.append(
effL1Muon3x3Truth.GetPaintedGraph().GetErrorYhigh(i) * 100)
yErr3x3TruthLow.append(
effL1Muon3x3Truth.GetPaintedGraph().GetErrorYlow(i) * 100)
xErrLow.append(0.25)
xErrHigh.append(.25)
for i in range(0, len(eff5x5)):
diff.append(effTruth3x3[i] - eff5x5[i])
diffErrHigh.append(
sqrt(pow(yErr3x3High[i], 2) + pow(yErr3x3TruthHigh[i], 2)))
diffErrLow.append(
sqrt(pow(yErr3x3Low[i], 2) + pow(yErr3x3TruthLow[i], 2)))
for i in range(1, len(xValues) - 1):
xErrLow.append((xValues[i] - xValues[i - 1]) / 2.)
xErrHigh.append((xValues[i + 1] - xValues[i]) / 2.)
xErrLow.append(xErrLow[-1])
xErrHigh.append(xErrHigh[-1])
fig = plt.figure()
ax1 = fig.add_subplot(211)
ax2 = fig.add_subplot(212, sharex=ax1)
ax1.errorbar(xValues,
eff5x5,
yerr=[yErr3x3Low, yErr3x3High],
xerr=[xErrLow, xErrHigh],
fmt='^',
color='#57AB27',
label='L1')
ax1.errorbar(xValues,
effTruth3x3,
yerr=[yErr3x3TruthLow, yErr3x3TruthHigh],
xerr=[xErrLow, xErrHigh],
fmt='^',
color='#E30066',
label='Truth L1')
ax1.legend(loc='lower right')
ax1.grid(True)
ax1.set_ylabel(r'$\epsilon$ / %')
ax1.set_title(
r'Efficiency $\epsilon$ for 3x3 grid in direct comparison')
pyplotCmsPrivateLabel(ax1)
ax2.errorbar(xValues,
diff,
yerr=[diffErrLow, diffErrHigh],
xerr=[xErrLow, xErrHigh],
fmt='o',
color='#00549F',
label=r'$\epsilon$(Truth L1) - $\epsilon$(L1)')
ax2.legend()
ax2.grid(True)
ax2.set_ylabel(r'$\Delta$ $\epsilon$ / %')
ax2.set_xlabel(r'p$_\mathrm{T}$ / GeV')
#make x labels of first plot invisible
plt.setp(ax1.get_xticklabels(), visible=False)
#avoid overlapping labels on y axis
plt.setp(ax1.get_yticklabels()[0], visible=False)
plt.setp(ax2.get_yticklabels()[-1], visible=False)
fig.subplots_adjust(hspace=0)
plt.savefig('plots/efficiency3x3.png')
plt.show(block=False)
def plotNtotalGridMatching3x3(self):
effL1Muon3x3 = self.fileHandler.getHistogram(
"efficiency/L1Muon3x3_Efficiency")
effL1Muon3x3Truth = self.fileHandler.getHistogram(
"efficiency/L1MuonTruth3x3_Efficiency")
effL1Muon5x5Truth = self.fileHandler.getHistogram(
"efficiency/L1MuonTruth5x5_Efficiency")
genPt = self.fileHandler.getHistogram("gen_Pt")
yTruth = []
yTruthErr = []
xValues = []
xErrLow = []
xErrHigh = []
yValues = []
xGen = []
xGenErr = []
yGen = []
yGenSummed = []
yGenSummedErr = []
for i in range(1, genPt.GetNbinsX()):
if (genPt.GetBinCenter(i) > 200):
break
xGen.append(genPt.GetBinCenter(i))
xGenErr.append(genPt.GetBinWidth(i) / 2.)
yGen.append(genPt.GetBinContent(i))
for i in range(1, effL1Muon3x3.GetPassedHistogram().GetNbinsX()):
if (effL1Muon3x3.GetTotalHistogram().GetBinContent(i) != 0):
yValues.append(
effL1Muon3x3.GetTotalHistogram().GetBinContent(i))
xValues.append(
effL1Muon3x3.GetTotalHistogram().GetBinCenter(i))
yTruth.append(
effL1Muon3x3Truth.GetTotalHistogram().GetBinContent(i))
yTruthErr.append(sqrt(yTruth[-1]))
print effL1Muon3x3.GetTotalHistogram().Integral(),effL1Muon3x3Truth.GetTotalHistogram().Integral()\
,genPt.Integral(),effL1Muon5x5Truth.GetTotalHistogram().Integral()
xErrLow.append(0.25)
xErrHigh.append(.25)
for i in range(1, len(xValues) - 1):
xErrLow.append((xValues[i] - xValues[i - 1]) / 2.)
xErrHigh.append((xValues[i + 1] - xValues[i]) / 2.)
xErrLow.append(xErrLow[-1])
xErrHigh.append(60) #xErrHigh[-1])
counter = 0
genSum = 0
for i in range(len(xGen)):
if (xGen[i] > (xValues[counter] + xErrHigh[counter])):
counter += 1
yGenSummed.append(genSum)
yGenSummedErr.append(sqrt(yGenSummed[-1]))
genSum = 0
genSum += yGen[i]
yGenSummed.append(genSum)
yGenSummedErr.append(sqrt(yGenSummed[-1]))
for i in range(0, len(yValues)):
yValues[i] /= (xErrHigh[i] + xErrLow[i])
for i in range(0, len(yTruth)):
yTruth[i] /= (xErrHigh[i] + xErrLow[i])
yTruthErr[i] /= (xErrHigh[i] + xErrLow[i])
for i in range(0, len(yGenSummed)):
yGenSummed[i] /= (xErrHigh[i] + xErrLow[i])
yGenSummedErr[i] /= (xErrHigh[i] + xErrLow[i])
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax1.errorbar(xValues,
yValues,
xerr=[xErrLow, xErrHigh],
fmt='^',
color=rwthDarkBlue,
label='L1')
ax1.set_yscale('log')
ax1.errorbar(xValues,
yTruth,
xerr=[xErrLow, xErrHigh],
yerr=yTruthErr,
fmt='^',
color=rwthGruen,
label='L1 Truth')
# ax1.errorbar(xGen,yGen,xerr=xGenErr,fmt='^',color=rwthMagenta,label='Gen')
ax1.errorbar(xValues,
yGenSummed,
xerr=[xErrLow, xErrHigh],
yerr=yGenSummedErr,
fmt='^',
color=rwthRot,
label='Gen Summed')
ax1.legend(loc='lower right')
ax1.grid(True)
ax1.set_ylabel('# / bin width (1/GeV)')
ax1.set_xlabel(r'p$_\mathrm{T}$ / GeV')
ax1.set_title('# L1')
pyplotCmsPrivateLabel(ax1)
plt.savefig('plots/nL1ByEmax3x3Absolute.png')
plt.show(block=False)
def plotL1TruthGridMatchingPlot(self):
effL1MuonCentral = self.fileHandler.getHistogram(
'efficiency/L1MuonTruthCentral_Efficiency')
effL1Muon3x3 = self.fileHandler.getHistogram(
'efficiency/L1MuonTruth3x3_Efficiency')
effL1Muon5x5 = self.fileHandler.getHistogram(
'efficiency/L1MuonTruth5x5_Efficiency')
c = TCanvas("c12", "c1", 1200, 1200)
effL1Muon5x5.SetMarkerColor(colorRwthMagenta)
effL1Muon5x5.SetLineColor(colorRwthMagenta)
effL1Muon5x5.SetMarkerStyle(27)
effL1Muon5x5.SetTitle(
'Efficiency for Matching HO to L1 truth;p_{T} / GeV;Efficiency')
effL1Muon5x5.Draw('')
c.Update()
effL1Muon5x5.GetPaintedGraph().GetYaxis().SetRangeUser(0.4, 1)
effL1Muon3x3.SetMarkerColor(colorRwthGruen)
effL1Muon3x3.SetLineColor(colorRwthGruen)
effL1Muon3x3.SetMarkerStyle(26)
effL1Muon3x3.Draw("same")
effL1MuonCentral.SetMarkerColor(colorRwthDarkBlue)
effL1MuonCentral.SetLineColor(colorRwthDarkBlue)
effL1MuonCentral.SetMarkerStyle(4)
effL1MuonCentral.Draw("same")
c.Update()
setupAxes(effL1MuonCentral)
setupAxes(effL1Muon3x3)
setupAxes(effL1Muon5x5)
legend = TLegend(0.55, 0.1, 0.9, 0.3)
legend.AddEntry(effL1MuonCentral, 'Matches in central tile', 'ep')
legend.AddEntry(effL1Muon3x3, 'Matches in 3x3 grid', 'ep')
legend.AddEntry(effL1Muon5x5, 'Matches in 5x5 grid', 'ep')
legend.Draw()
effCentral = []
eff3x3 = []
eff5x5 = []
xValues = []
xErrLow = []
xErrHigh = []
yErrCentralHigh = []
yErrCentralLow = []
yErr3x3High = []
yErr3x3Low = []
yErr5x5High = []
yErr5x5Low = []
for i in range(effL1MuonCentral.GetPassedHistogram().GetNbinsX()):
if effL1MuonCentral.GetTotalHistogram().GetBinContent(i) != 0:
effCentral.append(
effL1MuonCentral.GetPassedHistogram().GetBinContent(i) /
effL1MuonCentral.GetTotalHistogram().GetBinContent(i) *
100)
xValues.append(
effL1MuonCentral.GetTotalHistogram().GetBinCenter(i))
if effL1Muon3x3.GetTotalHistogram().GetBinContent(i) != 0:
eff3x3.append(
effL1Muon3x3.GetPassedHistogram().GetBinContent(i) /
effL1Muon3x3.GetTotalHistogram().GetBinContent(i) * 100)
if effL1Muon5x5.GetTotalHistogram().GetBinContent(i) != 0:
eff5x5.append(
effL1Muon5x5.GetPassedHistogram().GetBinContent(i) /
effL1Muon5x5.GetTotalHistogram().GetBinContent(i) * 100)
for i in range(effL1MuonCentral.GetPaintedGraph().GetN()):
yErrCentralHigh.append(
effL1MuonCentral.GetPaintedGraph().GetErrorYhigh(i) * 100)
yErrCentralLow.append(
effL1MuonCentral.GetPaintedGraph().GetErrorYlow(i) * 100)
yErr3x3High.append(
effL1Muon3x3.GetPaintedGraph().GetErrorYhigh(i) * 100)
yErr3x3Low.append(effL1Muon3x3.GetPaintedGraph().GetErrorYlow(i) *
100)
yErr5x5High.append(
effL1Muon5x5.GetPaintedGraph().GetErrorYhigh(i) * 100)
yErr5x5Low.append(effL1Muon5x5.GetPaintedGraph().GetErrorYlow(i) *
100)
xErrLow.append(0.25)
xErrHigh.append(.25)
for i in range(1, len(xValues) - 1):
xErrLow.append((xValues[i] - xValues[i - 1]) / 2.)
xErrHigh.append((xValues[i + 1] - xValues[i]) / 2.)
xErrLow.append(xErrLow[-1])
xErrHigh.append(xErrHigh[-1])
fig = plt.figure()
ax = fig.add_subplot(1, 1, 1)
plt.errorbar(xValues,
effCentral,
yerr=[yErrCentralLow, yErrCentralHigh],
xerr=[xErrLow, xErrHigh],
fmt='s',
color='#00549F',
label='Match in central tile')
plt.errorbar(xValues,
eff3x3,
yerr=[yErr3x3Low, yErr3x3High],
xerr=[xErrLow, xErrHigh],
fmt='o',
color='#E30066',
label='Match in 3x3')
plt.errorbar(xValues,
eff5x5,
yerr=[yErr5x5Low, yErr5x5High],
xerr=[xErrLow, xErrHigh],
fmt='^',
color='#57AB27',
label='Match in 5x5')
plt.gca().xaxis.grid(True)
plt.gca().yaxis.grid(True)
plt.axis([0., 155, 40, 101])
plt.xlabel(r'p$_{\mathrm{T}}$ / GeV')
plt.ylabel(r'$\epsilon$ / %')
plt.title(r'Efficiency $\epsilon$ of matching "true" L1 to HO')
plt.legend(loc='lower right')
pyplotCmsPrivateLabel(ax)
plt.savefig('plots/efficiencyTruth.png')
plt.show(block=False)
