def main():
set_root_defaults()
options, _ = parse_arguments()
variable = 'ST'
config_7TeV = XSectionConfig(7)
config_8TeV = XSectionConfig(8)
path_to_JSON_7TeV = options.path + '/7TeV/' + variable + '/'
path_to_JSON_8TeV = options.path + '/8TeV/' + variable + '/'
# we need the generators
# and the central samples + errors
results_7TeV, _ = read_xsection_measurement_results( path_to_JSON_7TeV,
variable,
bin_edges,
category = 'central',
channel = 'combined',
k_values = {
'combined': config_7TeV.k_values_combined}
)
results_8TeV, _ = read_xsection_measurement_results( path_to_JSON_8TeV,
variable,
bin_edges,
category = 'central',
channel = 'combined',
k_values = {
'combined': config_8TeV.k_values_combined}
)
plot_results(results_7TeV, results_8TeV, variable)
for _, histogram_dict in histograms.iteritems():
for _, histogram in histogram_dict.iteritems():
histogram = asrootpy(histogram)
histogram.Rebin(rebin)
histogram.Scale(scale_factor)
def get_title(variable):
if '3jets' in variable:
return 'CMS Preliminary, $\mathcal{L}$ = 5.0 fb$^{-1}$ at $\sqrt{s}$ = 7 TeV \n e+jets, 3 jets'
else:
return 'CMS Preliminary, $\mathcal{L}$ = 5.0 fb$^{-1}$ at $\sqrt{s}$ = 7 TeV \n e+jets, $\geq$4 jets'
if __name__ == '__main__':
set_root_defaults()
CMS.title['fontsize'] = 40
CMS.x_axis_title['fontsize'] = 50
CMS.y_axis_title['fontsize'] = 50
CMS.axis_label_major['labelsize'] = 40
CMS.axis_label_minor['labelsize'] = 40
CMS.legend_properties['size'] = 40
# for reliso
rebin = 10
path_to_files = '/storage/TopQuarkGroup/results/histogramfiles/AN-11-265_V1/'
lumi = 1959.75
data = 'data'
pfmuon = ''
histogram_files = {
'TTJet': path_to_files + 'TTJet_%spb_PFElectron_%sPF2PATJets_PFMET.root' % (str(lumi), pfmuon),
'matchingup': matchingup_normalised_xsection,
'scaledown': scaledown_normalised_xsection,
'scaleup': scaleup_normalised_xsection
}
if not channel == 'combined':
filename = path_to_JSON + '/xsection_measurement_results/%s/kv%d/%s/normalised_xsection_%s.txt' % ( channel, k_value, category, met_type )
else:
filename = path_to_JSON + '/xsection_measurement_results/%s/%s/normalised_xsection_%s.txt' % ( channel, category, met_type )
if normalise_to_one:
filename = filename.replace( 'normalised_xsection', 'normalised_to_one_xsection' )
write_data_to_JSON( normalised_xsection, filename )
if __name__ == '__main__':
set_root_defaults( msg_ignore_level = 3001 )
# setup
parser = OptionParser()
parser.add_option( "-p", "--path", dest = "path", default = 'data/',
help = "set path to JSON files" )
parser.add_option( "-v", "--variable", dest = "variable", default = 'MET',
help = "set the variable to analyse (MET, HT, ST, MT)" )
parser.add_option( "-b", "--bjetbin", dest = "bjetbin", default = '2m',
help = "set b-jet multiplicity for analysis. Options: exclusive: 0-3, inclusive (N or more): 0m, 1m, 2m, 3m, 4m" )
parser.add_option( "-m", "--metType", dest = "metType", default = 'type1',
help = "set MET type for analysis of MET, ST or MT" )
parser.add_option( "-f", "--load_fakes", dest = "load_fakes", action = "store_true",
help = "Load fakes histogram and perform manual fake subtraction in TSVDUnfold" )
parser.add_option( "-u", "--unfolding_method", dest = "unfolding_method", default = 'RooUnfoldSvd',
help = "Unfolding method: RooUnfoldSvd (default), TSVDUnfold, TopSVDUnfold, RooUnfoldTUnfold, RooUnfoldInvert, RooUnfoldBinByBin, RooUnfoldBayes" )
parser.add_option( "-H", "--hreco", type = 'int',
fakes.SetColor('red')
h_fakes.SetColor('blue')
fakes.SetFillStyle('\\')
h_fakes.SetFillStyle('/')
#cross check: are the fakes the same?
plt.figure(figsize=(16, 10), dpi=100)
rplt.hist(fakes, label=r'fakes from unfolding', stacked=False)
rplt.hist(h_fakes, label=r'fakes from MC', stacked=False, alpha = 0.5)
plt.xlabel('$E_{\mathrm{T}}^{miss}$')
plt.ylabel('Events')
plt.title('Unfolding')
plt.legend()
plt.savefig('plots/Fakes' + outputfile_suffix + '.png')
if __name__ == "__main__":
set_root_defaults()
method = 'RooUnfoldSvd' # = 'RooUnfoldBayes | RooUnfoldSvd | RooUnfoldBinByBin | RooUnfoldInvert | RooUnfoldTUnfold
bins = array('d', [0, 25, 45, 70, 100, 1000])
nbins = len(bins) - 1
inputFile = File('../data/unfolding_merged_sub1.root', 'read')
h_truth = asrootpy(inputFile.unfoldingAnalyserElectronChannel.truth.Rebin(nbins, 'truth', bins))
h_measured = asrootpy(inputFile.unfoldingAnalyserElectronChannel.measured.Rebin(nbins, 'measured', bins))
h_fakes = asrootpy(inputFile.unfoldingAnalyserElectronChannel.fake.Rebin(nbins, 'fake', bins))
h_response = inputFile.unfoldingAnalyserElectronChannel.response_withoutFakes_AsymBins #response_AsymBins
# h_measured_new = h_measured - h_fakes
# h_response = inputFile.unfoldingAnalyserElectronChannel.response_AsymBins #response_AsymBins
nEvents = inputFile.EventFilter.EventCounter.GetBinContent(1)
lumiweight = 164.5 * 5050 / nEvents
h_truth.Scale(lumiweight)
def main():
"Main Function"
set_root_defaults()
parser = OptionParser("Script to check progress of CRAB jobs in creating nTuples. Run as: python check_CRAB_jobs.py -p projectFolder -n numberOfJobs >&check.log &")
parser.add_option("-p", "--projectFolder", dest="projectFolder", help="specify project")
parser.add_option("-n", "--numberOfJobs", dest="numberOfJobs",
help="specify project")
(options, _) = parser.parse_args()
#make sure the project option has been specified
if not options.projectFolder:
parser.error('Please enter a project folder as the -p option: /gpfs_phys/storm/cms/user/...')
#normalise the projectFolder filepath and add a "/" at the end
projectFolder = os.path.normpath(options.projectFolder) + os.sep
#list the items in the CRAB output folder on the Bristol Storage Element.
storageElementList=glob.glob(projectFolder + "*.root")
if storageElementList:
pass
else:
print "Location Error: Specified project folder does not exist on the Bristol Storage Element, signifying that the CRAB job has probably not started running yet or you forgot to include the full path /gpfs_storm/cms/user/..."
sys.exit()
#The following section has been commented out because if it is the first time this script is being run in a session, a grid password will be needed which will cause the script
#to not be able to finish. Since the only purpose of this following CRAB command is to obtain the number of jobs, for the time being the number of jobs has been entered as an option to
#the script which should be manually entered by the user.
#get the status of the crab jobs and extract the number of output files expected on the Bristol Storage Element.
# projectFolder = options.projectFolder.split("/")[6]
# status = commands.getstatusoutput("crab -status -c " + projectFolder)
# statusFormatted = status[1].split("\n")
# for line in statusFormatted:
# if "crab:" in line and "Total Jobs" in line:
# words = line.split()
# numberOfJobs = int(words[1])
#Now, check that all job root files are present in Bristol Storage Element folder:
missingOrBrokenTemp = []
missingOrBroken = []
goodFilesTemp = []
goodFiles = []
presentJobList = []
duplicatesToDelete = []
#make list of all the job numbers which should be present.
jobList = range(1,int(options.numberOfJobs)+1)
#try opening all files in Bristol Storage Element folder and add to missing list if they cannot be opened.
for f in storageElementList:
#make list of all jobs numbers in the Bristol Storage Element folder
jobNumber = int((re.split('[\W+,_]',f))[-4])
presentJobList.append(jobNumber)
#check if files are corrupt or not
try:
rootFile = File(f)
rootFile.Close()
except:
print "Adding Job Number", jobNumber, "to missingOrBroken list because file is corrupted."
missingOrBrokenTemp.append(jobNumber)
else:
goodFilesTemp.append(jobNumber)
#now add any absent files to the missing list:
for job in jobList:
if job not in presentJobList:
print "Adding Job Number", job, "to missingOrBroken list because it doesn't exist on the Storage Element."
missingOrBrokenTemp.append(job)
#Remove any job numbers from missingOrBroken which appear in both goodFiles and missingOrBroken lists
for job in missingOrBrokenTemp:
if job not in goodFilesTemp:
missingOrBroken.append(job)
else:
print "Removing", job, "from missingOrBroken list because there is at least one duplicate good output file."
#Remove any job numbers from goodFiles which appear more than once in goodFiles
for job in goodFilesTemp:
if job not in goodFiles:
goodFiles.append(job)
else:
duplicatesToDelete.append(job)
print "\n The following", len(goodFiles), "good output files were found in the Bristol Storage Element folder:"
print str(goodFiles).replace(" ", "")
print "\n The following", len(duplicatesToDelete), "job numbers have multiple good files on the Bristol Storage Element folder which can be deleted:"
print str(duplicatesToDelete).replace(" ", "")
print "\n The following", len(missingOrBroken), "job numbers could not be found in the Bristol Storage Element folder:"
print str(missingOrBroken).replace(" ", "")
def main():
set_root_defaults()
# prevent directory ownership of ROOT histograms (python does the garbage collection)
TH1F.AddDirectory( False )
parser = OptionParser()
parser.add_option( "-n", "--n_toy_mc",
dest = "n_toy_mc", default = 300,
help = "number of toy MC to create", type = int )
parser.add_option( "-o", "--output",
dest = "output_folder", default = 'data/toy_mc/',
help = "output folder for toy MC" )
parser.add_option( "-v", "--variable", dest = "variable", default = 'MET',
help = "set the variable to analyse (MET, HT, ST, MT, WPT)" )
parser.add_option( "-m", "--metType", dest = "metType", default = 'type1',
help = "set MET type for analysis of MET, ST or MT" )
parser.add_option( "-c", "--centre-of-mass-energy", dest = "CoM", default = 8,
help = "set the centre of mass energy for analysis. Default = 8 [TeV]", type = int )
parser.add_option( '-V', '--verbose', dest = "verbose", action = "store_true",
help = "Print the event number, reco and gen variable value" )
( options, _ ) = parser.parse_args()
measurement_config = XSectionConfig( options.CoM )
centre_of_mass = options.CoM
ttbar_xsection = measurement_config.ttbar_xsection
variable = options.variable
met_type = measurement_config.translate_options[options.metType]
n_toy_mc = options.n_toy_mc
make_folder_if_not_exists( options.output_folder )
# get histograms
input_file_hists = File( measurement_config.unfolding_madgraph )
# define output file
out_file_template = '%s/toy_mc_%s_N_%d_%dTeV.root'
out_file_name = out_file_template % (options.output_folder, variable, n_toy_mc, centre_of_mass)
output = File( out_file_name, 'recreate' )
for channel in ['electron', 'muon']:
# first get the weights
h_truth, h_measured, h_response, _ = get_unfold_histogram_tuple( input_file_hists,
variable,
channel,
met_type,
centre_of_mass,
ttbar_xsection,
load_fakes = False )
# create directories
directory = output.mkdir( channel )
mkdir = directory.mkdir
cd = directory.cd
cd()
# generate toy MC
for i in range( 1, n_toy_mc + 1 ):
mkdir( 'toy_%d' % i )
cd( 'toy_%d' % i )
# create histograms
# add tuples (truth, measured, response) of histograms
truth = generate_toy_MC_from_distribution(h_truth)
measured = generate_toy_MC_from_distribution(h_measured)
response = generate_toy_MC_from_2Ddistribution(h_response)
truth.SetName('truth')
measured.SetName('measured')
response.SetName('response')
truth.Write()
measured.Write()
response.Write()
output.Write()
output.Close()
