def getNormFactorFileList(dirName, fileBaseName):
scriptList = []
# For-loop: All items (files/dir) in directory
for item in os.listdir(dirName):
fullPath = os.path.join(dirName, item)
# Skip directories
if os.path.isdir(fullPath):
continue
# Find files matching the script "Base" name (without moduleInfoStrings)
if item.startswith((fileBaseName).replace("%s.py", "")):
if item.endswith(".py"):
scriptList.append(item)
if len(scriptList) < 1:
msg = "ERROR! Found no normalization info files under dir %s. Did you generate them?" % dirName
raise Exception(ShellStyles.ErrorStyle() + msg +
ShellStyles.NormalStyle())
else:
msg = "Found %s norm-factor file(s):\n\t%s" % (
len(scriptList), "\n\t".join(
os.path.join([os.path.join(dirName, s) for s in scriptList])))
Verbose(ShellStyles.NoteLabel() + msg, True)
return scriptList
default=NORM_EWK_SRC,
help="Source of EWK normalisation [default: %s" %
(NORM_EWK_SRC))
(opts, parseArgs) = parser.parse_args()
# Require at least two arguments (script-name, path to multicrab)
if len(sys.argv) < 2:
parser.print_help()
sys.exit(1)
if opts.mcrab == None:
Print(
"Not enough arguments passed to script execution. Printing docstring & EXIT."
)
parser.print_help()
#print __doc__
sys.exit(1)
if opts.useInclusiveNorm:
msg = "Will use only inclusive weight instead of binning (no splitted histograms)"
Print(ShellStyles.NoteLabel() + msg, True)
# Call the main function
main()
if not opts.batchMode:
raw_input(
"=== makeInvertedPseudoMulticrab.py: Press any key to quit ROOT ..."
)
def importNormFactors(era, searchMode, optimizationMode, multicrabDirName):
'''
Imports the auto-generates QCDInvertedNormalizationFactors.py file, which is
created by the plotting/fitting templates script (plotQCD_Fit.py)
This containsthe results of fitting to the Baseline Data the templates m_{jjb}
shapes from the QCD (Inverted Data) and EWK (Baseline MC).
Results include the fit details for each shape and the QCD NormFactor for moving
from the ControlRegion (CR) to the Signal Region (SR).
The aforementioned python file and a folder with the histogram ROOT files and the individual
fits. The foler name will be normalisationPlots/<OptsMode> and will be placed inside the
<pseudomulticrab_dir>. The autogenerated file file be place in the cwd (i.e. work/)
'''
# Find candidates for normalisation scripts
scriptList = getNormFactorFileList(dirName=multicrabDirName,
fileBaseName=opts.normFactorsSrc)
# Create a string with the module information used
moduleInfoString = getModuleInfoString(era, searchMode, optimizationMode)
# Construct source file name
src = getGetNormFactorsSrcFilename(multicrabDirName,
opts.normFactorsSrc % moduleInfoString)
# Check if normalization coefficients are suitable for the choses era
Verbose("Reading normalisation factors from:\n\t%s" % src, True)
# Split the path to get just the file name of src
pathList = src.replace(".py", "").split("/")
# Insert the directory where the normFactor files reside into the path so that they are found
if len(pathList) > 1:
cwd = os.getenv("PWD")
# Get directories to src in a list [i.e. remove the last entry (file-name) from the pathList]
dirList = map(str, pathList[:(len(pathList) - 1)])
srcDir = "/".join(dirList)
sys.path.insert(0, os.path.join(cwd, srcDir))
# Import the (normFactor) src file
normFactorsImport = __import__(os.path.basename("/".join(pathList)))
# Get the function definition
myNormFactorsSafetyCheck = getattr(normFactorsImport,
"QCDInvertedNormalizationSafetyCheck")
Verbose(
"Check that the era=%s, searchMode=%s, optimizationMode=%s info matches!"
% (era, searchMode, optimizationMode))
myNormFactorsSafetyCheck(era, searchMode, optimizationMode)
# Obtain normalization factors
myNormFactorsImport = getattr(normFactorsImport, "QCDNormalization")
msg = "Disabled NormFactors Syst Var Fake Weighting Up/Down"
Print(ShellStyles.WarningLabel() + msg, True)
# myNormFactorsImportSystVarFakeWeightingDown = getattr(normFactorsImport, "QCDPlusEWKFakeTausNormalizationSystFakeWeightingVarDown") #FIXME
# myNormFactorsImportSystVarFakeWeightingUp = getattr(normFactorsImport, "QCDPlusEWKFakeTausNormalizationSystFakeWeightingVarUp") #FIXME
myNormFactors = {}
myNormFactors["nominal"] = myNormFactorsImport
msg = "Obtained \"nominal\" QCD normalisation factors dictionary. The values are:\n"
for k in myNormFactors["nominal"]:
msg += "\t" + k + " = " + str(myNormFactors["nominal"][k])
Print(ShellStyles.NoteLabel() + msg, True)
msg = "Disabled NormFactors Weighting Up/Down"
Print(ShellStyles.WarningLabel() + msg, True)
# myNormFactors["FakeWeightingDown"] = myNormFactorsImportSystVarFakeWeightingDown # FIXME
# myNormFactors["FakeWeightingUp"] = myNormFactorsImportSystVarFakeWeightingUp # FIXME
return myNormFactors
def _doCalculate(self, shape, moduleInfoString, normFactors, optionPrintPurityByBins, optionDoNQCDByBinHistograms):
'''
Calculates the result
'''
Verbose("Calculate final shape in signal region (shape * w_QCD) & initialize result containers", True)
nSplitBins = shape.getNumberOfPhaseSpaceSplitBins()
Verbose("Create Shape", True)
self._resultShape = aux.Clone(shape.getDataDrivenQCDHistoForSplittedBin(0))
self._resultShape.Reset()
self._resultShape.SetTitle("NQCDFinal_Total_%s"%moduleInfoString)
self._resultShape.SetName("NQCDFinal_Total_%s"%moduleInfoString)
Verbose("Create EWK shape", True)
self._resultShapeEWK = aux.Clone(shape.getDataDrivenQCDHistoForSplittedBin(0))
self._resultShapeEWK.Reset()
self._resultShapeEWK.SetTitle("NQCDFinal_EWK_%s"%moduleInfoString)
self._resultShapeEWK.SetName("NQCDFinal_EWK_%s"%moduleInfoString)
Verbose("Create Purity shape", True)
self._resultShapePurity = aux.Clone(shape.getDataDrivenQCDHistoForSplittedBin(0))
self._resultShapePurity.Reset()
self._resultShapePurity.SetTitle("NQCDFinal_Purity_%s"%moduleInfoString)
self._resultShapePurity.SetName("NQCDFinal_Purity_%s"%moduleInfoString)
self._histogramsList = []
myUncertaintyLabels = ["statData", "statEWK"]
self._resultCountObject = extendedCount.ExtendedCount(0.0, [0.0, 0.0], myUncertaintyLabels)
if optionDoNQCDByBinHistograms:
for i in range(0, nSplitBins):
hBin = aux.Clone(self._resultShape)
hBin.SetTitle("NQCDFinal_%s_%s"%(shape.getPhaseSpaceBinFileFriendlyTitle(i).replace(" ",""), moduleInfoString))
hBin.SetName("NQCDFinal_%s_%s"%(shape.getPhaseSpaceBinFileFriendlyTitle(i).replace(" ",""), moduleInfoString))
self._histogramsList.append(hBin)
if isinstance(self._resultShape, ROOT.TH2):
self._doCalculate2D(nSplitBins, shape, normFactors, optionPrintPurityByBins, optionDoNQCDByBinHistograms, myUncertaintyLabels)
return
# Intialize counters for purity calculation in final shape binning
myShapeDataSum = []
myShapeDataSumUncert = []
myShapeEwkSum = []
myShapeEwkSumUncert = []
# For-loop: All Bins
for j in range(1,self._resultShape.GetNbinsX()+1):
myShapeDataSum.append(0.0)
myShapeDataSumUncert.append(0.0)
myShapeEwkSum.append(0.0)
myShapeEwkSumUncert.append(0.0)
Verbose("Calculate results separately for each phase-space bin and then combine", True)
# For-loop: All measurement bins (e.g. tau pT bins for HToTauNu)
for i in range(0, nSplitBins):
# N.B: The \"Inclusive\" value is in the zeroth bin
Verbose("Get data-driven QCD, data, and MC EWK shape histogram for the phase-space bin", True)
h = shape.getDataDrivenQCDHistoForSplittedBin(i)
hData = shape.getDataHistoForSplittedBin(i)
hEwk = shape.getEwkHistoForSplittedBin(i)
Verbose("Get normalization factor", True)
wQCDLabel = shape.getPhaseSpaceBinFileFriendlyTitle(i)
if self._optionUseInclusiveNorm:
wQCDLabel = "Inclusive"
wQCD = 0.0
if not wQCDLabel in normFactors.keys():
msg = "No normalization factors available for bin '%s' when accessing histogram %s! Ignoring this bin..." % (wQCDLabel,shape.getHistoName())
Print(ShellStyles.WarningLabel() + msg, True)
else:
wQCD = normFactors[wQCDLabel]
msg = "Weighting bin \"%i\" (label=\"%s\") with normFactor \"%s\"" % (i, wQCDLabel, wQCD)
Verbose(ShellStyles.NoteLabel() + msg, True)
# Construct info table (debugging)
table = []
align = "{:>6} {:^10} {:^15} {:>10} {:>10} {:>10} {:^3} {:^8} {:^3} {:^8}"
header = align.format("Bin", "Width", "Range", "Content", "NormFactor", "QCD", "+/-", "Data", "+/-", "EWK")
hLine = "="*90
table.append("{:^90}".format(shape.getHistoName()))
table.append(hLine)
table.append(header)
table.append(hLine)
binSum = 0.0
nBins = h.GetNbinsX()
binWidth = hData.GetBinWidth(0)
xMin = hData.GetXaxis().GetBinCenter(0)
xMax = hData.GetXaxis().GetBinCenter(nBins+1)
# For-Loop (nested): All bins in the shape histogram
for j in range(1, nBins+1):
# Initialise values
myResult = 0.0
myStatDataUncert = 0.0
myStatEwkUncert = 0.0
# Ignore zero bins
if abs(h.GetBinContent(j)) > 0.00001:
Verbose("Calculating the result")
binContent = h.GetBinContent(j)
binRange = "%.1f -> %.1f" % (h.GetXaxis().GetBinLowEdge(j), h.GetXaxis().GetBinUpEdge(j) )
binWidth = GetTH1BinWidthString(h, j)
binSum += binContent
myResult = binContent * wQCD #apply normalisation factor (transfer from CR to SR))
Verbose("Calculate abs. stat. uncert. for data and for MC EWK (Do not calculate here MC EWK syst.)", True)
myStatDataUncert = hData.GetBinError(j) * wQCD
myStatEwkUncert = hEwk.GetBinError(j) * wQCD
table.append(align.format(j, binWidth, binRange, "%0.1f" % binContent, wQCD, "%.1f" % myResult, "+/-", "%.1f" % myStatDataUncert, "+/-", "%.1f" % myStatEwkUncert))
# Get count object
myCountObject = extendedCount.ExtendedCount(myResult, [myStatDataUncert, myStatEwkUncert], myUncertaintyLabels)
self._resultCountObject.add(myCountObject)
if optionDoNQCDByBinHistograms:
Verbose("Setting bin content \"%i\"" % (j), True)
self._histogramsList[i].SetBinContent(j, myCountObject.value())
self._histogramsList[i].SetBinError(j, myCountObject.statUncertainty())
binContent = self._resultShape.GetBinContent(j) + myCountObject.value()
binError = self._resultShape.GetBinError(j) + myCountObject.statUncertainty()**2
Verbose("Setting bin %i to content %0.1f +/- %0.1f" % (j, binContent, binError), j==0)
self._resultShape.SetBinContent(j, binContent)
self._resultShape.SetBinError(j, binError) # Sum squared (take sqrt outside loop on final squared sum)
Verbose("Sum items for purity calculation", True)
myShapeDataSum[j-1] += hData.GetBinContent(j)*wQCD
myShapeDataSumUncert[j-1] += (hData.GetBinError(j)*wQCD)**2
myShapeEwkSum[j-1] += hEwk.GetBinContent(j)*wQCD
myShapeEwkSumUncert[j-1] += (hEwk.GetBinError(j)*wQCD)**2
# Delete the shape histograms
h.Delete()
hData.Delete()
hEwk.Delete()
# For-loop: All shape bins
for j in range(1,self._resultShape.GetNbinsX()+1):
# Take square root of uncertainties
self._resultShape.SetBinError(j, math.sqrt(self._resultShape.GetBinError(j)))
# Print detailed results in a formatted table
qcdResults = self._resultCountObject.getResultAndStatErrorsDict()
bins = "%0.f-%.0f" % (1, nBins)
binRange = "%.1f -> %.1f" % (xMin, xMax)
binSum = "%.1f" % binSum
nQCD = "%.1f" % qcdResults["value"]
dataStat = "%.1f" % qcdResults["statData"]
ewkStat = "%.1f" % qcdResults["statEWK"]
table.append(align.format(bins, binWidth, binRange, binSum, wQCD, nQCD, "+/-", dataStat, "+/-", ewkStat))
table.append(hLine)
for i, line in enumerate(table):
if i == len(table)-2:
Verbose(ShellStyles.TestPassedStyle()+line+ShellStyles.NormalStyle(), i==0)
else:
Verbose(line, i==0)
if optionPrintPurityByBins:
Verbose("Printing Shape Purity bin-by-bin.", True)
self.PrintPurityByBins(nBins, shape, myShapeDataSum, myShapeDataSumUncert, myShapeEwkSum, myShapeEwkSumUncert)
return
