| 1 |
#!/usr/bin/env python
|
| 2 |
import sys
|
| 3 |
import os
|
| 4 |
import re
|
| 5 |
import time
|
| 6 |
from math import *
|
| 7 |
from array import *
|
| 8 |
from decimal import *
|
| 9 |
from optparse import OptionParser
|
| 10 |
from OSUT3Analysis.Configuration.configurationOptions import *
|
| 11 |
from OSUT3Analysis.Configuration.processingUtilities import *
|
| 12 |
from OSUT3Analysis.Configuration.formattingUtilities import *
|
| 13 |
|
| 14 |
|
| 15 |
|
| 16 |
### parse the command-line options
|
| 17 |
|
| 18 |
parser = OptionParser()
|
| 19 |
parser = set_commandline_arguments(parser)
|
| 20 |
|
| 21 |
parser.add_option("-f", "--fancy", action="store_true", dest="makeFancy", default=False,
|
| 22 |
help="removes the title and replaces it with the official CMS plot heading")
|
| 23 |
parser.add_option("--ylog", action="store_true", dest="setLogY", default=False,
|
| 24 |
help="Set logarithmic scale on vertical axis on all plots")
|
| 25 |
parser.add_option("--ymin", dest="setYMin",
|
| 26 |
help="Minimum of y axis")
|
| 27 |
parser.add_option("--ymax", dest="setYMax",
|
| 28 |
help="Maximum of y axis")
|
| 29 |
parser.add_option("-E", "--ratioRelErrMax", dest="ratioRelErrMax",
|
| 30 |
help="maximum error used in rebinning the ratio histogram")
|
| 31 |
|
| 32 |
(arguments, args) = parser.parse_args()
|
| 33 |
|
| 34 |
|
| 35 |
if arguments.localConfig:
|
| 36 |
sys.path.append(os.getcwd())
|
| 37 |
exec("from " + arguments.localConfig.rstrip('.py') + " import *")
|
| 38 |
|
| 39 |
#### deal with conflicting arguments
|
| 40 |
if arguments.normalizeToData and arguments.normalizeToUnitArea:
|
| 41 |
print "Conflicting normalizations requsted, will normalize to unit area"
|
| 42 |
arguments.normalizeToData = False
|
| 43 |
if arguments.normalizeToData and arguments.noStack:
|
| 44 |
print "You have asked to scale non-stacked backgrounds to data. This is a very strange request. Will normalize to unit area instead"
|
| 45 |
arguments.normalizeToData = False
|
| 46 |
arguments.normalizeToUnitArea = True
|
| 47 |
if arguments.makeRatioPlots and arguments.makeDiffPlots:
|
| 48 |
print "You have requested both ratio and difference plots. Will make just ratio plots instead"
|
| 49 |
arguments.makeRatioPlots = False
|
| 50 |
if arguments.makeRatioPlots and arguments.noStack:
|
| 51 |
print "You have asked to make a ratio plot and to not stack the backgrounds. This is a very strange request. Will skip making the ratio plot."
|
| 52 |
arguments.makeRatioPlots = False
|
| 53 |
if arguments.makeDiffPlots and arguments.noStack:
|
| 54 |
print "You have asked to make a difference plot and to not stack the backgrounds. This is a very strange request. Will skip making the difference plot."
|
| 55 |
arguments.makeDiffPlots = False
|
| 56 |
|
| 57 |
|
| 58 |
from ROOT import TFile, gROOT, gStyle, gDirectory, TStyle, THStack, TH1F, TCanvas, TString, TLegend, TLegendEntry, THStack, TIter, TKey, TPaveLabel, TPaveText, TF1, gPad
|
| 59 |
|
| 60 |
|
| 61 |
### setting ROOT options so our plots will look awesome and everyone will love us
|
| 62 |
|
| 63 |
gROOT.SetBatch()
|
| 64 |
gStyle.SetOptStat(0)
|
| 65 |
gStyle.SetCanvasBorderMode(0)
|
| 66 |
gStyle.SetPadBorderMode(0)
|
| 67 |
gStyle.SetPadColor(0)
|
| 68 |
gStyle.SetCanvasColor(0)
|
| 69 |
gStyle.SetTextFont(42)
|
| 70 |
gStyle.SetCanvasDefH(600)
|
| 71 |
gStyle.SetCanvasDefW(600)
|
| 72 |
gStyle.SetCanvasDefX(0)
|
| 73 |
gStyle.SetCanvasDefY(0)
|
| 74 |
gStyle.SetPadTopMargin(0.07)
|
| 75 |
gStyle.SetPadBottomMargin(0.13)
|
| 76 |
gStyle.SetPadLeftMargin(0.15)
|
| 77 |
gStyle.SetPadRightMargin(0.05)
|
| 78 |
gStyle.SetTitleColor(1, "XYZ")
|
| 79 |
gStyle.SetTitleFont(42, "XYZ")
|
| 80 |
gStyle.SetTitleSize(0.04, "XYZ")
|
| 81 |
gStyle.SetTitleXOffset(1.1)
|
| 82 |
gStyle.SetTitleYOffset(2)
|
| 83 |
gStyle.SetTextAlign(12)
|
| 84 |
gStyle.SetLabelColor(1, "XYZ")
|
| 85 |
gStyle.SetLabelFont(42, "XYZ")
|
| 86 |
gStyle.SetLabelOffset(0.007, "XYZ")
|
| 87 |
gStyle.SetLabelSize(0.04, "XYZ")
|
| 88 |
gStyle.SetAxisColor(1, "XYZ")
|
| 89 |
gStyle.SetStripDecimals(True)
|
| 90 |
gStyle.SetTickLength(0.03, "XYZ")
|
| 91 |
gStyle.SetNdivisions(510, "XYZ")
|
| 92 |
gStyle.SetPadTickX(1)
|
| 93 |
gStyle.SetPadTickY(1)
|
| 94 |
gROOT.ForceStyle()
|
| 95 |
|
| 96 |
|
| 97 |
#set the text for the luminosity label
|
| 98 |
if(intLumi < 1000.):
|
| 99 |
LumiInPb = intLumi
|
| 100 |
LumiText = "L_{int} = " + str(intLumi) + " pb^{-1}"
|
| 101 |
LumiText = "L_{int} = " + str.format('{0:.1f}', LumiInPb) + " pb^{-1}"
|
| 102 |
else:
|
| 103 |
LumiInFb = intLumi/1000.
|
| 104 |
LumiText = "L_{int} = " + str.format('{0:.1f}', LumiInFb) + " fb^{-1}"
|
| 105 |
|
| 106 |
#bestest place for lumi. label, in top left corner
|
| 107 |
topLeft_x_left = 0.1375839
|
| 108 |
topLeft_x_right = 0.4580537
|
| 109 |
topLeft_y_bottom = 0.8479021
|
| 110 |
topLeft_y_top = 0.9475524
|
| 111 |
topLeft_y_offset = 0.035
|
| 112 |
|
| 113 |
#set the text for the fancy heading
|
| 114 |
HeaderText = "CMS Preliminary: " + LumiText + " at #sqrt{s} = 8 TeV"
|
| 115 |
|
| 116 |
#position for header
|
| 117 |
header_x_left = 0.2181208
|
| 118 |
header_x_right = 0.9562937
|
| 119 |
header_y_bottom = 0.9479866
|
| 120 |
header_y_top = 0.9947552
|
| 121 |
|
| 122 |
|
| 123 |
|
| 124 |
##########################################################################################################################################
|
| 125 |
##########################################################################################################################################
|
| 126 |
##########################################################################################################################################
|
| 127 |
|
| 128 |
# some fancy-ass code from Andrzej Zuranski to merge bins in the ratio plot until the error goes below some threshold
|
| 129 |
def ratioHistogram( dataHist, mcHist, relErrMax=0.10):
|
| 130 |
|
| 131 |
if not dataHist:
|
| 132 |
print "Error: trying to run ratioHistogram but dataHist is invalid"
|
| 133 |
return
|
| 134 |
|
| 135 |
if not mcHist:
|
| 136 |
print "Error: trying to run ratioHistogram but mcHist is invalid"
|
| 137 |
return
|
| 138 |
|
| 139 |
def groupR(group):
|
| 140 |
Data,MC = [float(sum(hist.GetBinContent(i) for i in group)) for hist in [dataHist,mcHist]]
|
| 141 |
return (Data-MC)/MC if MC else 0
|
| 142 |
|
| 143 |
def groupErr(group):
|
| 144 |
Data,MC = [float(sum(hist.GetBinContent(i) for i in group)) for hist in [dataHist,mcHist]]
|
| 145 |
dataErr2,mcErr2 = [sum(hist.GetBinError(i)**2 for i in group) for hist in [dataHist,mcHist]]
|
| 146 |
if Data > 0 and MC > 0 and Data != MC:
|
| 147 |
return abs(math.sqrt( (dataErr2+mcErr2)/(Data-MC)**2 + mcErr2/MC**2 ) * (Data-MC)/MC)
|
| 148 |
else:
|
| 149 |
return 0
|
| 150 |
|
| 151 |
def regroup(groups):
|
| 152 |
err,iG = max( (groupErr(g),groups.index(g)) for g in groups )
|
| 153 |
if err < relErrMax or len(groups)<3 : return groups
|
| 154 |
iH = max( [iG-1,iG+1], key = lambda i: groupErr(groups[i]) if 0<=i<len(groups) else -1 )
|
| 155 |
iLo,iHi = sorted([iG,iH])
|
| 156 |
return regroup(groups[:iLo] + [groups[iLo]+groups[iHi]] + groups[iHi+1:])
|
| 157 |
|
| 158 |
#don't rebin the histograms of the number of a given object (except for the pileup ones)
|
| 159 |
if ((dataHist.GetName().find("num") is not -1 and dataHist.GetName().find("Primaryvertexs") is -1) or
|
| 160 |
dataHist.GetName().find("CutFlow") is not -1 or
|
| 161 |
dataHist.GetName().find("GenMatch") is not -1):
|
| 162 |
ratio = dataHist.Clone()
|
| 163 |
ratio.Add(mcHist,-1)
|
| 164 |
ratio.Divide(mcHist)
|
| 165 |
ratio.SetTitle("")
|
| 166 |
else:
|
| 167 |
groups = regroup( [(i,) for i in range(1,1+dataHist.GetNbinsX())] )
|
| 168 |
ratio = TH1F("ratio","",len(groups), array('d', [dataHist.GetBinLowEdge(min(g)) for g in groups ] + [dataHist.GetXaxis().GetBinUpEdge(dataHist.GetNbinsX())]) )
|
| 169 |
for i,g in enumerate(groups) :
|
| 170 |
ratio.SetBinContent(i+1,groupR(g))
|
| 171 |
ratio.SetBinError(i+1,groupErr(g))
|
| 172 |
|
| 173 |
ratio.GetYaxis().SetTitle("#frac{Data-MC}{MC}")
|
| 174 |
ratio.SetLineColor(1)
|
| 175 |
ratio.SetLineWidth(2)
|
| 176 |
return ratio
|
| 177 |
|
| 178 |
##########################################################################################################################################
|
| 179 |
##########################################################################################################################################
|
| 180 |
##########################################################################################################################################
|
| 181 |
|
| 182 |
|
| 183 |
def MakeOneDHist(pathToDir,distribution):
|
| 184 |
|
| 185 |
numFittingSamples = 0
|
| 186 |
|
| 187 |
HeaderLabel = TPaveLabel(header_x_left,header_y_bottom,header_x_right,header_y_top,HeaderText,"NDC")
|
| 188 |
HeaderLabel.SetTextAlign(32)
|
| 189 |
HeaderLabel.SetBorderSize(0)
|
| 190 |
HeaderLabel.SetFillColor(0)
|
| 191 |
HeaderLabel.SetFillStyle(0)
|
| 192 |
|
| 193 |
LumiLabel = TPaveLabel(topLeft_x_left,topLeft_y_bottom,topLeft_x_right,topLeft_y_top,LumiText,"NDC")
|
| 194 |
LumiLabel.SetBorderSize(0)
|
| 195 |
LumiLabel.SetFillColor(0)
|
| 196 |
LumiLabel.SetFillStyle(0)
|
| 197 |
|
| 198 |
NormLabel = TPaveLabel()
|
| 199 |
NormLabel.SetDrawOption("NDC")
|
| 200 |
NormLabel.SetX1NDC(topLeft_x_left)
|
| 201 |
NormLabel.SetX2NDC(topLeft_x_right)
|
| 202 |
|
| 203 |
NormLabel.SetBorderSize(0)
|
| 204 |
NormLabel.SetFillColor(0)
|
| 205 |
NormLabel.SetFillStyle(0)
|
| 206 |
|
| 207 |
NormText = ""
|
| 208 |
if arguments.normalizeToUnitArea:
|
| 209 |
NormText = "Scaled to unit area"
|
| 210 |
elif arguments.normalizeToData:
|
| 211 |
NormText = "MC scaled to data"
|
| 212 |
NormLabel.SetLabel(NormText)
|
| 213 |
|
| 214 |
YieldsLabel = TPaveText(0.39, 0.7, 0.59, 0.9,"NDC")
|
| 215 |
YieldsLabel.SetBorderSize(0)
|
| 216 |
YieldsLabel.SetFillColor(0)
|
| 217 |
YieldsLabel.SetFillStyle(0)
|
| 218 |
YieldsLabel.SetTextAlign(12)
|
| 219 |
|
| 220 |
RatiosLabel = TPaveText()
|
| 221 |
RatiosLabel.SetDrawOption("NDC")
|
| 222 |
RatiosLabel.SetBorderSize(0)
|
| 223 |
RatiosLabel.SetFillColor(0)
|
| 224 |
RatiosLabel.SetFillStyle(0)
|
| 225 |
RatiosLabel.SetTextAlign(32)
|
| 226 |
|
| 227 |
|
| 228 |
Legend = TLegend()
|
| 229 |
Legend.SetBorderSize(0)
|
| 230 |
Legend.SetFillColor(0)
|
| 231 |
Legend.SetFillStyle(0)
|
| 232 |
|
| 233 |
|
| 234 |
fittingIntegral = 0
|
| 235 |
scaleFactor = 1
|
| 236 |
|
| 237 |
HistogramsToFit = []
|
| 238 |
TargetDataset = distribution['target_dataset']
|
| 239 |
|
| 240 |
FittingLegendEntries = []
|
| 241 |
DataLegendEntries = []
|
| 242 |
|
| 243 |
FittingHistogramDatasets = []
|
| 244 |
|
| 245 |
|
| 246 |
Stack_list = []
|
| 247 |
Stack_list.append (THStack("stack_before",distribution['name']))
|
| 248 |
Stack_list.append (THStack("stack_after",distribution['name']))
|
| 249 |
|
| 250 |
fileName = condor_dir + "/" + distribution['target_dataset'] + ".root"
|
| 251 |
if not os.path.exists(fileName):
|
| 252 |
return
|
| 253 |
inputFile = TFile(fileName)
|
| 254 |
if inputFile.IsZombie() or not inputFile.GetNkeys():
|
| 255 |
return
|
| 256 |
|
| 257 |
|
| 258 |
|
| 259 |
Target = inputFile.Get("OSUAnalysis/"+distribution['channel']+"/"+distribution['name']).Clone()
|
| 260 |
Target.SetDirectory(0)
|
| 261 |
inputFile.Close()
|
| 262 |
|
| 263 |
Target.SetMarkerStyle(20)
|
| 264 |
Target.SetMarkerSize(0.8)
|
| 265 |
Target.SetFillStyle(0)
|
| 266 |
Target.SetLineColor(colors[TargetDataset])
|
| 267 |
Target.SetLineStyle(1)
|
| 268 |
Target.SetLineWidth(2)
|
| 269 |
targetIntegral = Target.Integral()
|
| 270 |
if(arguments.normalizeToUnitArea and Target.Integral() > 0):
|
| 271 |
Target.Scale(1./Target.Integral())
|
| 272 |
|
| 273 |
### formatting target histogram and adding to legend
|
| 274 |
legendIndex = 0
|
| 275 |
Legend.AddEntry(Target,labels[TargetDataset],"LEP")
|
| 276 |
legendIndex = legendIndex+1
|
| 277 |
|
| 278 |
if not outputFile.Get ("OSUAnalysis"):
|
| 279 |
outputFile.mkdir ("OSUAnalysis")
|
| 280 |
if not outputFile.Get ("OSUAnalysis/" + distribution['channel']):
|
| 281 |
outputFile.Get ("OSUAnalysis").mkdir (distribution['channel'])
|
| 282 |
|
| 283 |
for sample in distribution['datasets']: # loop over different samples requested to be fit
|
| 284 |
|
| 285 |
dataset_file = "%s/%s.root" % (condor_dir,sample)
|
| 286 |
inputFile = TFile(dataset_file)
|
| 287 |
HistogramObj = inputFile.Get(pathToDir+"/"+distribution['channel']+"/"+distribution['name'])
|
| 288 |
if not HistogramObj:
|
| 289 |
print "WARNING: Could not find histogram " + pathToDir + "/" + distribution['name'] + " in file " + dataset_file + ". Will skip it and continue."
|
| 290 |
continue
|
| 291 |
Histogram = HistogramObj.Clone()
|
| 292 |
Histogram.SetDirectory(0)
|
| 293 |
inputFile.Close()
|
| 294 |
if arguments.rebinFactor:
|
| 295 |
RebinFactor = int(arguments.rebinFactor)
|
| 296 |
#don't rebin histograms which will have less than 5 bins or any gen-matching histograms
|
| 297 |
if Histogram.GetNbinsX() >= RebinFactor*5 and Histogram.GetName().find("GenMatch") is -1:
|
| 298 |
Histogram.Rebin(RebinFactor)
|
| 299 |
|
| 300 |
xAxisLabel = Histogram.GetXaxis().GetTitle()
|
| 301 |
unitBeginIndex = xAxisLabel.find("[")
|
| 302 |
unitEndIndex = xAxisLabel.find("]")
|
| 303 |
|
| 304 |
if unitBeginIndex is not -1 and unitEndIndex is not -1: #x axis has a unit
|
| 305 |
yAxisLabel = "Entries / " + str(Histogram.GetXaxis().GetBinWidth(1)) + " " + xAxisLabel[unitBeginIndex+1:unitEndIndex]
|
| 306 |
else:
|
| 307 |
yAxisLabel = "Entries per bin (" + str(Histogram.GetXaxis().GetBinWidth(1)) + " width)"
|
| 308 |
|
| 309 |
if not arguments.makeFancy:
|
| 310 |
histoTitle = Histogram.GetTitle()
|
| 311 |
else:
|
| 312 |
histoTitle = ""
|
| 313 |
|
| 314 |
|
| 315 |
legLabel = labels[sample]
|
| 316 |
if (arguments.printYields):
|
| 317 |
yieldHist = Histogram.Integral()
|
| 318 |
legLabel = legLabel + " (%.1f)" % yieldHist
|
| 319 |
FittingLegendEntries.append(legLabel)
|
| 320 |
|
| 321 |
if( types[sample] == "bgMC"):
|
| 322 |
|
| 323 |
numFittingSamples += 1
|
| 324 |
fittingIntegral += Histogram.Integral()
|
| 325 |
|
| 326 |
Histogram.SetLineStyle(1)
|
| 327 |
if(arguments.noStack):
|
| 328 |
Histogram.SetFillStyle(0)
|
| 329 |
Histogram.SetLineColor(colors[sample])
|
| 330 |
Histogram.SetLineWidth(2)
|
| 331 |
else:
|
| 332 |
Histogram.SetFillStyle(1001)
|
| 333 |
Histogram.SetFillColor(colors[sample])
|
| 334 |
Histogram.SetLineColor(1)
|
| 335 |
Histogram.SetLineWidth(1)
|
| 336 |
|
| 337 |
elif( types[sample] == "signalMC"):
|
| 338 |
|
| 339 |
numFittingSamples += 1
|
| 340 |
|
| 341 |
Histogram.SetFillStyle(0)
|
| 342 |
Histogram.SetLineColor(colors[sample])
|
| 343 |
Histogram.SetLineStyle(1)
|
| 344 |
Histogram.SetLineWidth(2)
|
| 345 |
if(arguments.normalizeToUnitArea and Histogram.Integral() > 0):
|
| 346 |
Histogram.Scale(1./Histogram.Integral())
|
| 347 |
|
| 348 |
HistogramsToFit.append(Histogram)
|
| 349 |
FittingHistogramDatasets.append(sample)
|
| 350 |
|
| 351 |
#scaling histograms as per user's specifications
|
| 352 |
if targetIntegral > 0 and fittingIntegral > 0:
|
| 353 |
scaleFactor = targetIntegral/fittingIntegral
|
| 354 |
for fittingHist in HistogramsToFit:
|
| 355 |
if arguments.normalizeToData:
|
| 356 |
fittingHist.Scale(scaleFactor)
|
| 357 |
|
| 358 |
if arguments.normalizeToUnitArea and not arguments.noStack and fittingIntegral > 0:
|
| 359 |
fittingHist.Scale(1./fittingIntegral)
|
| 360 |
elif arguments.normalizeToUnitArea and arguments.noStack and fittingHist.Integral() > 0:
|
| 361 |
fittingHist.Scale(1./fittingHist.Integral())
|
| 362 |
|
| 363 |
|
| 364 |
def fitf (x, par):
|
| 365 |
xBin = HistogramsToFit[0].FindBin (x[0])
|
| 366 |
value = 0.0
|
| 367 |
sumOfWeights = 0.0
|
| 368 |
|
| 369 |
for i in range (0, len (HistogramsToFit)):
|
| 370 |
error = HistogramsToFit[i].GetBinError (xBin)
|
| 371 |
if error != 0.0:
|
| 372 |
weight = 1.0 / (error * error)
|
| 373 |
sumOfWeights += weight
|
| 374 |
value += weight * par[i] * HistogramsToFit[i].GetBinContent (xBin)
|
| 375 |
if sumOfWeights != 0.0:
|
| 376 |
value /= sumOfWeights
|
| 377 |
|
| 378 |
return value
|
| 379 |
|
| 380 |
|
| 381 |
def fitf (x, par):
|
| 382 |
xBin = HistogramsToFit[0].FindBin (x[0])
|
| 383 |
value = 0.0
|
| 384 |
|
| 385 |
for i in range (0, len (HistogramsToFit)):
|
| 386 |
value += par[i] * HistogramsToFit[i].GetBinContent (xBin)
|
| 387 |
|
| 388 |
return value
|
| 389 |
|
| 390 |
|
| 391 |
lowerLimit = Target.GetBinLowEdge (1)
|
| 392 |
upperLimit = Target.GetBinLowEdge (Target.GetNbinsX ()) + Target.GetBinWidth (Target.GetNbinsX ())
|
| 393 |
if 'lowerLimit' in distribution:
|
| 394 |
lowerLimit = distribution['lowerLimit']
|
| 395 |
if 'upperLimit' in distribution:
|
| 396 |
upperLimit = distribution['upperLimit']
|
| 397 |
func = TF1 ("fit", fitf, lowerLimit, upperLimit, len (HistogramsToFit))
|
| 398 |
|
| 399 |
for i in range (0, len (HistogramsToFit)):
|
| 400 |
if 'fixed_datasets' in distribution and distribution['datasets'][i] in distribution['fixed_datasets']:
|
| 401 |
func.FixParameter (i, 1.0)
|
| 402 |
else:
|
| 403 |
func.SetParameter (i, 1.0)
|
| 404 |
func.SetParName (i, labels[FittingHistogramDatasets[i]])
|
| 405 |
|
| 406 |
for i in range (0, distribution['iterations'] - 1):
|
| 407 |
print "Iteration " + str (i + 1) + "..."
|
| 408 |
Target.Fit ("fit", "QEMR0")
|
| 409 |
Target.Fit ("fit", "VEMR0")
|
| 410 |
|
| 411 |
|
| 412 |
finalMax = 0
|
| 413 |
if not arguments.noStack:
|
| 414 |
for fittingHist in HistogramsToFit:
|
| 415 |
finalMax += fittingHist.GetMaximum()
|
| 416 |
else:
|
| 417 |
for fittingHist in HistogramsToFit:
|
| 418 |
if(fittingHist.GetMaximum() > finalMax):
|
| 419 |
finalMax = fittingHist.GetMaximum()
|
| 420 |
if(Target.GetMaximum() > finalMax):
|
| 421 |
finalMax = Target.GetMaximum()
|
| 422 |
|
| 423 |
Target.SetMaximum(1.1*finalMax)
|
| 424 |
Target.SetMinimum(0.0001)
|
| 425 |
|
| 426 |
Canvas = TCanvas(distribution['name'] + "_FitFunction")
|
| 427 |
Canvas.cd (1)
|
| 428 |
Target.Draw ()
|
| 429 |
func.Draw ("same")
|
| 430 |
|
| 431 |
outputFile.cd ("OSUAnalysis/" + distribution['channel'])
|
| 432 |
Canvas.Write ()
|
| 433 |
if arguments.savePDFs:
|
| 434 |
if histogram == input_histograms[0]:
|
| 435 |
Canvas.Print (pdfFileName + "(", "pdf")
|
| 436 |
else:
|
| 437 |
Canvas.Print (pdfFileName, "pdf")
|
| 438 |
Target.SetStats (0)
|
| 439 |
|
| 440 |
|
| 441 |
|
| 442 |
|
| 443 |
### formatting bgMC histograms and adding to legend
|
| 444 |
legendIndex = numFittingSamples-1
|
| 445 |
for Histogram in reversed(HistogramsToFit):
|
| 446 |
if(arguments.noStack):
|
| 447 |
Legend.AddEntry(Histogram,FittingLegendEntries[legendIndex],"L")
|
| 448 |
else:
|
| 449 |
Legend.AddEntry(Histogram,FittingLegendEntries[legendIndex],"F")
|
| 450 |
legendIndex = legendIndex-1
|
| 451 |
|
| 452 |
|
| 453 |
### Drawing histograms to canvas
|
| 454 |
|
| 455 |
makeRatioPlots = arguments.makeRatioPlots
|
| 456 |
makeDiffPlots = arguments.makeDiffPlots
|
| 457 |
|
| 458 |
yAxisMin = 0.0001
|
| 459 |
if arguments.setYMin:
|
| 460 |
yAxisMin = float(arguments.setYMin)
|
| 461 |
|
| 462 |
|
| 463 |
### Draw everything to the canvases !!!!
|
| 464 |
|
| 465 |
for i in range (0, 2): # 0 => before, 1 => after
|
| 466 |
|
| 467 |
if i == 1:
|
| 468 |
ratios = []
|
| 469 |
errors = []
|
| 470 |
for j in range (0, len (HistogramsToFit)):
|
| 471 |
HistogramsToFit[j].Scale (func.GetParameter (j))
|
| 472 |
ratios.append(func.GetParameter (j))
|
| 473 |
errors.append(func.GetParError(j))
|
| 474 |
|
| 475 |
for fittingHist in HistogramsToFit:
|
| 476 |
if not arguments.noStack:
|
| 477 |
Stack_list[i].Add(fittingHist)
|
| 478 |
|
| 479 |
|
| 480 |
#creating the histogram to represent the statistical errors on the stack
|
| 481 |
if not arguments.noStack:
|
| 482 |
ErrorHisto = HistogramsToFit[0].Clone("errors")
|
| 483 |
ErrorHisto.SetFillStyle(3001)
|
| 484 |
ErrorHisto.SetFillColor(13)
|
| 485 |
ErrorHisto.SetLineWidth(0)
|
| 486 |
if i == 1:
|
| 487 |
Legend.AddEntry(ErrorHisto,"Stat. Errors","F")
|
| 488 |
for Histogram in HistogramsToFit:
|
| 489 |
if Histogram is not HistogramsToFit[0]:
|
| 490 |
ErrorHisto.Add(Histogram)
|
| 491 |
|
| 492 |
if i == 0:
|
| 493 |
Canvas = TCanvas(distribution['name'] + "_Before")
|
| 494 |
if i == 1:
|
| 495 |
Canvas = TCanvas(distribution['name'] + "_After")
|
| 496 |
|
| 497 |
|
| 498 |
if makeRatioPlots or makeDiffPlots:
|
| 499 |
Canvas.SetFillStyle(0)
|
| 500 |
Canvas.Divide(1,2)
|
| 501 |
Canvas.cd(1)
|
| 502 |
gPad.SetPad(0,0.25,1,1)
|
| 503 |
gPad.SetMargin(0.15,0.05,0.01,0.07)
|
| 504 |
gPad.SetFillStyle(0)
|
| 505 |
gPad.Update()
|
| 506 |
gPad.Draw()
|
| 507 |
if arguments.setLogY:
|
| 508 |
gPad.SetLogy()
|
| 509 |
Canvas.cd(2)
|
| 510 |
gPad.SetPad(0,0,1,0.25)
|
| 511 |
# format: gPad.SetMargin(l,r,b,t)
|
| 512 |
gPad.SetMargin(0.15,0.05,0.4,0.01)
|
| 513 |
gPad.SetFillStyle(0)
|
| 514 |
gPad.SetGridy(1)
|
| 515 |
gPad.Update()
|
| 516 |
gPad.Draw()
|
| 517 |
|
| 518 |
Canvas.cd(1)
|
| 519 |
|
| 520 |
### finding the maximum value of anything going on the canvas, so we know how to set the y-axis
|
| 521 |
finalMax = 0
|
| 522 |
if numFittingSamples is not 0 and not arguments.noStack:
|
| 523 |
finalMax = ErrorHisto.GetMaximum() + ErrorHisto.GetBinError(ErrorHisto.GetMaximumBin())
|
| 524 |
else:
|
| 525 |
for bgMCHist in HistogramsToFit:
|
| 526 |
if(bgMCHist.GetMaximum() > finalMax):
|
| 527 |
finalMax = bgMCHist.GetMaximum()
|
| 528 |
if(Target.GetMaximum() > finalMax):
|
| 529 |
finalMax = Target.GetMaximum() + Target.GetBinError(Target.GetMaximumBin())
|
| 530 |
finalMax = 1.15*finalMax
|
| 531 |
if arguments.setYMax:
|
| 532 |
finalMax = float(arguments.setYMax)
|
| 533 |
|
| 534 |
|
| 535 |
if not arguments.noStack: # draw stacked background samples
|
| 536 |
Stack_list[i].SetTitle(histoTitle)
|
| 537 |
Stack_list[i].Draw("HIST")
|
| 538 |
Stack_list[i].GetXaxis().SetTitle(xAxisLabel)
|
| 539 |
Stack_list[i].GetYaxis().SetTitle(yAxisLabel)
|
| 540 |
Stack_list[i].SetMaximum(finalMax)
|
| 541 |
Stack_list[i].SetMinimum(yAxisMin)
|
| 542 |
if makeRatioPlots or makeDiffPlots:
|
| 543 |
Stack_list[i].GetHistogram().GetXaxis().SetLabelSize(0)
|
| 544 |
#draw shaded error bands
|
| 545 |
ErrorHisto.Draw("A E2 SAME")
|
| 546 |
|
| 547 |
else: #draw the unstacked backgrounds
|
| 548 |
HistogramsToFit[0].SetTitle(histoTitle)
|
| 549 |
HistogramsToFit[0].Draw("HIST")
|
| 550 |
HistogramsToFit[0].GetXaxis().SetTitle(xAxisLabel)
|
| 551 |
HistogramsToFit[0].GetYaxis().SetTitle(yAxisLabel)
|
| 552 |
HistogramsToFit[0].SetMaximum(finalMax)
|
| 553 |
HistogramsToFit[0].SetMinimum(yAxisMin)
|
| 554 |
for bgMCHist in HistogramsToFit:
|
| 555 |
bgMCHist.Draw("A HIST SAME")
|
| 556 |
|
| 557 |
Target.Draw("A E X0 SAME")
|
| 558 |
|
| 559 |
|
| 560 |
|
| 561 |
#legend coordinates, empirically determined :-)
|
| 562 |
x_left = 0.6761745
|
| 563 |
x_right = 0.9328859
|
| 564 |
x_width = x_right - x_left
|
| 565 |
y_max = 0.9
|
| 566 |
entry_height = 0.05
|
| 567 |
|
| 568 |
if(numFittingSamples is not 0): #then draw the data & bgMC legend
|
| 569 |
|
| 570 |
numExtraEntries = 2 # count the target and (lack of) title
|
| 571 |
Legend.SetX1NDC(x_left)
|
| 572 |
numExtraEntries = numExtraEntries + 1 # count the stat. errors entry
|
| 573 |
|
| 574 |
Legend.SetY1NDC(y_max-entry_height*(numExtraEntries+numFittingSamples))
|
| 575 |
Legend.SetX2NDC(x_right)
|
| 576 |
Legend.SetY2NDC(y_max)
|
| 577 |
Legend.Draw()
|
| 578 |
|
| 579 |
RatiosLabel.SetX1NDC(x_left - 0.1)
|
| 580 |
RatiosLabel.SetX2NDC(x_right)
|
| 581 |
RatiosLabel.SetY2NDC(Legend.GetY1NDC() - 0.1)
|
| 582 |
RatiosLabel.SetY1NDC(RatiosLabel.GetY2NDC() - entry_height*(numFittingSamples))
|
| 583 |
|
| 584 |
# Deciding which text labels to draw and drawing them
|
| 585 |
drawLumiLabel = False
|
| 586 |
drawNormLabel = False
|
| 587 |
offsetNormLabel = False
|
| 588 |
drawHeaderLabel = False
|
| 589 |
|
| 590 |
if not arguments.normalizeToUnitArea: #don't draw the lumi label if there's no data and it's scaled to unit area
|
| 591 |
drawLumiLabel = True
|
| 592 |
# move the normalization label down before drawing if we drew the lumi. label
|
| 593 |
offsetNormLabel = True
|
| 594 |
if arguments.normalizeToUnitArea or arguments.normalizeToData:
|
| 595 |
drawNormLabel = True
|
| 596 |
if arguments.makeFancy:
|
| 597 |
drawHeaderLabel = True
|
| 598 |
drawLumiLabel = False
|
| 599 |
|
| 600 |
# now that flags are set, draw the appropriate labels
|
| 601 |
|
| 602 |
if drawLumiLabel:
|
| 603 |
LumiLabel.Draw()
|
| 604 |
|
| 605 |
if drawNormLabel:
|
| 606 |
if offsetNormLabel:
|
| 607 |
NormLabel.SetY1NDC(topLeft_y_bottom-topLeft_y_offset)
|
| 608 |
NormLabel.SetY2NDC(topLeft_y_top-topLeft_y_offset)
|
| 609 |
else:
|
| 610 |
NormLabel.SetY1NDC(topLeft_y_bottom)
|
| 611 |
NormLabel.SetY2NDC(topLeft_y_top)
|
| 612 |
NormLabel.Draw()
|
| 613 |
|
| 614 |
if drawHeaderLabel:
|
| 615 |
HeaderLabel.Draw()
|
| 616 |
|
| 617 |
YieldsLabel.Clear()
|
| 618 |
mcYield = Stack_list[i].GetStack().Last().Integral()
|
| 619 |
dataYield = Target.Integral()
|
| 620 |
if i == 0:
|
| 621 |
YieldsLabel.AddText ("Before Fit to Data")
|
| 622 |
if i == 1:
|
| 623 |
YieldsLabel.AddText ("After Fit to Data")
|
| 624 |
YieldsLabel.AddText ("data yield: " + '%.1f' % dataYield)
|
| 625 |
YieldsLabel.AddText ("MC yield: " + '%.1f' % mcYield)
|
| 626 |
if i == 1:
|
| 627 |
for j in range(0,len(FittingLegendEntries)):
|
| 628 |
RatiosLabel.AddText (FittingLegendEntries[j]+" ratio: " + '%.2f' % ratios[j] + ' #pm %.2f' % errors[j])
|
| 629 |
YieldsLabel.Draw()
|
| 630 |
RatiosLabel.Draw()
|
| 631 |
|
| 632 |
# drawing the ratio or difference plot if requested
|
| 633 |
if (makeRatioPlots or makeDiffPlots):
|
| 634 |
Canvas.cd(2)
|
| 635 |
BgSum = Stack_list[i].GetStack().Last()
|
| 636 |
if makeRatioPlots:
|
| 637 |
if arguments.ratioRelErrMax:
|
| 638 |
Comparison = ratioHistogram(Target,BgSum,arguments.ratioRelErrMax)
|
| 639 |
else:
|
| 640 |
Comparison = ratioHistogram(Target,BgSum)
|
| 641 |
elif makeDiffPlots:
|
| 642 |
Comparison = Target.Clone("diff")
|
| 643 |
Comparison.Add(BgSum,-1)
|
| 644 |
Comparison.SetTitle("")
|
| 645 |
Comparison.GetYaxis().SetTitle("Data-MC")
|
| 646 |
Comparison.GetXaxis().SetTitle(xAxisLabel)
|
| 647 |
Comparison.GetYaxis().CenterTitle()
|
| 648 |
Comparison.GetYaxis().SetTitleSize(0.1)
|
| 649 |
Comparison.GetYaxis().SetTitleOffset(0.5)
|
| 650 |
Comparison.GetXaxis().SetTitleSize(0.15)
|
| 651 |
Comparison.GetYaxis().SetLabelSize(0.1)
|
| 652 |
Comparison.GetXaxis().SetLabelSize(0.15)
|
| 653 |
if makeRatioPlots:
|
| 654 |
RatioYRange = 1.15
|
| 655 |
if arguments.ratioYRange:
|
| 656 |
RatioYRange = float(arguments.ratioYRange)
|
| 657 |
Comparison.GetYaxis().SetRangeUser(-1*RatioYRange, RatioYRange)
|
| 658 |
elif makeDiffPlots:
|
| 659 |
YMax = Comparison.GetMaximum()
|
| 660 |
YMin = Comparison.GetMinimum()
|
| 661 |
if YMax <= 0 and YMin <= 0:
|
| 662 |
Comparison.GetYaxis().SetRangeUser(-1.2*YMin,0)
|
| 663 |
elif YMax >= 0 and YMin >= 0:
|
| 664 |
Comparison.GetYaxis().SetRangeUser(0,1.2*YMax)
|
| 665 |
else: #axis crosses y=0
|
| 666 |
if abs(YMax) > abs(YMin):
|
| 667 |
Comparison.GetYaxis().SetRangeUser(-1.2*YMax,1.2*YMax)
|
| 668 |
else:
|
| 669 |
Comparison.GetYaxis().SetRangeUser(-1.2*YMin,1.2*YMin)
|
| 670 |
|
| 671 |
Comparison.GetYaxis().SetNdivisions(205)
|
| 672 |
Comparison.Draw()
|
| 673 |
|
| 674 |
|
| 675 |
|
| 676 |
if i == 0:
|
| 677 |
Canvas.Write (distribution['name'] + "_Before")
|
| 678 |
if arguments.savePDFs:
|
| 679 |
pathToDirString = plainTextString(pathToDir)
|
| 680 |
Canvas.SaveAs(condor_dir+"/fitting_histogram_pdfs/"+pathToDirString+"/"+distribution['name']+"_Before.pdf")
|
| 681 |
|
| 682 |
if i == 1:
|
| 683 |
Canvas.Write (distribution['name'] + "_After")
|
| 684 |
if arguments.savePDFs:
|
| 685 |
pathToDirString = plainTextString(pathToDir)
|
| 686 |
Canvas.SaveAs(condor_dir+"/fitting_histogram_pdfs/"+pathToDirString+"/"+distribution['name']+"_After.pdf")
|
| 687 |
|
| 688 |
|
| 689 |
|
| 690 |
|
| 691 |
##########################################################################################################################################
|
| 692 |
##########################################################################################################################################
|
| 693 |
##########################################################################################################################################
|
| 694 |
|
| 695 |
|
| 696 |
##########################################################################################################################################
|
| 697 |
##########################################################################################################################################
|
| 698 |
##########################################################################################################################################
|
| 699 |
|
| 700 |
|
| 701 |
condor_dir = set_condor_output_dir(arguments)
|
| 702 |
|
| 703 |
|
| 704 |
# make output file
|
| 705 |
outputFileName = "mc_fit_to_data.root"
|
| 706 |
if arguments.outputFileName:
|
| 707 |
outputFileName = arguments.outputFileName
|
| 708 |
|
| 709 |
outputFile = TFile(condor_dir + "/" + outputFileName, "RECREATE")
|
| 710 |
|
| 711 |
|
| 712 |
if arguments.savePDFs:
|
| 713 |
os.system("rm -rf %s/fitting_histograms_pdfs" % (condor_dir))
|
| 714 |
os.system("mkdir %s/fitting_histograms_pdfs" % (condor_dir))
|
| 715 |
|
| 716 |
|
| 717 |
#get root directory in the first layer, generally "OSUAnalysis"
|
| 718 |
for distribution in input_distributions:
|
| 719 |
MakeOneDHist("OSUAnalysis",distribution)
|
| 720 |
|
| 721 |
outputFile.Close()
|
