def test_tau_closure(self):
for channel in self.channels:
for variable in self.variables:
data = self.dict[channel][variable]["h_measured"]
truth = hist_to_value_error_tuplelist(self.dict[channel][variable]["h_truth"])
unfolded_result = hist_to_value_error_tuplelist(
self.dict[channel][variable]["tau_unfolding_object"].unfold(data)
)
# the difference between the truth and unfolded result should
# be within the unfolding error
for (value, error), (true_value, _) in zip(unfolded_result, truth):
self.assertAlmostEquals(value, true_value, delta=error)
def test_tau_closure(self):
for channel in self.channels:
for variable in self.variables:
data = self.dict[channel][variable]['h_measured']
truth = hist_to_value_error_tuplelist(
self.dict[channel][variable]['h_truth'])
unfolded_result = hist_to_value_error_tuplelist(
self.dict[channel][variable]
['tau_unfolding_object'].unfold(data))
# the difference between the truth and unfolded result should
# be within the unfolding error
for (value, error), (true_value,
_) in zip(unfolded_result, truth):
self.assertAlmostEquals(value, true_value, delta=error)
def __init__(self,
config,
measurement,
method=BACKGROUND_SUBTRACTION,
phase_space='FullPS'):
self.config = config
self.variable = measurement.variable
self.category = measurement.name
self.channel = measurement.channel
self.method = method
self.phase_space = phase_space
self.measurement = measurement
self.measurement.read()
self.met_type = measurement.met_type
self.fit_variables = ['M3']
self.normalisation = {}
self.initial_normalisation = {}
self.templates = {}
self.have_normalisation = False
for sample, hist in self.measurement.histograms.items():
h = deepcopy(hist)
h_norm = h.integral()
if h_norm > 0:
h.Scale(1 / h.integral())
self.templates[sample] = hist_to_value_error_tuplelist(h)
self.auxiliary_info = {}
self.auxiliary_info['norms'] = measurement.aux_info_norms
def calculate_normalisation(self):
'''
1. get file names
2. get histograms from files
3. ???
4. calculate normalisation
'''
if self.have_normalisation:
return
histograms = self.measurement.histograms
for sample, hist in histograms.items():
# TODO: this should be a list of bin-contents
hist = fix_overflow(hist)
histograms[sample] = hist
self.initial_normalisation[sample] = hist_to_value_error_tuplelist(
hist)
self.normalisation[sample] = self.initial_normalisation[sample]
self.background_subtraction(histograms)
# next, let's round all numbers (they are event numbers after all
for sample, values in self.normalisation.items():
new_values = [(round(v, 1), round(e, 1)) for v, e in values]
self.normalisation[sample] = new_values
self.have_normalisation = True
def calculate_normalisation(self):
'''
1. get file names
2. get histograms from files
3. ???
4. calculate normalisation based on self.method
'''
if self.have_normalisation:
return
histograms = self.measurement.histograms
for sample, hist in histograms.items():
# TODO: this should be a list of bin-contents
hist = fix_overflow(hist)
histograms[sample] = hist
self.initial_normalisation[
sample] = hist_to_value_error_tuplelist(hist)
if self.method == self.BACKGROUND_SUBTRACTION and sample != 'TTJet':
self.normalisation[sample] = self.initial_normalisation[sample]
if self.method == self.BACKGROUND_SUBTRACTION:
self.background_subtraction(histograms)
if self.method == self.SIMULTANEOUS_FIT:
self.simultaneous_fit(histograms)
# next, let's round all numbers (they are event numbers after all
for sample, values in self.normalisation.items():
new_values = [(round(v, 1), round(e, 1)) for v, e in values]
self.normalisation[sample] = new_values
self.have_normalisation = True
def calculate_normalisation(self):
'''
1. get file names
2. get histograms from files
3. ???
4. calculate normalisation
'''
if self.have_normalisation:
return
histograms = self.measurement.histograms
for sample, hist in histograms.items():
# TODO: this should be a list of bin-contents
hist = fix_overflow(hist)
histograms[sample] = hist
self.initial_normalisation[sample] = hist_to_value_error_tuplelist(hist)
self.normalisation[sample] = self.initial_normalisation[sample]
self.background_subtraction(histograms)
# next, let's round all numbers (they are event numbers after all
for sample, values in self.normalisation.items():
new_values = [(round(v, 1), round(e, 1)) for v, e in values]
self.normalisation[sample] = new_values
self.have_normalisation = True
def unfold_results( results, category, channel, tau_value, h_truth, h_measured, h_response, h_fakes, method, visiblePS ):
global variable, path_to_DF, args
edges = reco_bin_edges_full[variable]
if visiblePS:
edges = reco_bin_edges_vis[variable]
h_data = value_error_tuplelist_to_hist( results, edges )
# Rebin original TTJet_Measured in terms of final binning (h_data is later replaced with h_data_no_fakes)
h_data_rebinned = h_data.rebinned(2)
# Remove fakes before unfolding
h_data_no_fakes = removeFakes( h_measured, h_fakes, h_data )
# unfold
unfolding = Unfolding( h_data_no_fakes, h_truth, h_measured, h_response, h_fakes, method = method, tau = tau_value )
# turning off the unfolding errors for systematic samples
if not category == 'central':
unfoldCfg.error_treatment = 0
else:
unfoldCfg.error_treatment = args.error_treatment
h_unfolded_data = unfolding.unfold()
h_data_no_fakes = h_data_no_fakes.rebinned(2)
covariance_matrix = None
if category == 'central':
# Return the covariance matrices (They have been normailsed)
covariance_matrix, correlation_matrix = unfolding.get_covariance_matrix()
# Write covariance matrices
covariance_output_tempalte = '{path_to_DF}/central/covarianceMatrices/{cat}_{label}_{channel}.txt'
# Unfolded number of events
table_outfile=covariance_output_tempalte.format( path_to_DF=path_to_DF, channel = channel, label='Covariance', cat='Stat_unfoldedNormalisation' )
create_covariance_matrix( covariance_matrix, table_outfile)
table_outfile=covariance_output_tempalte.format( path_to_DF=path_to_DF, channel = channel, label='Correlation', cat='Stat_unfoldedNormalisation' )
create_covariance_matrix( correlation_matrix, table_outfile )
# # Normalised cross section
# table_outfile=covariance_output_tempalte.format( path_to_DF=path_to_DF, channel = channel, label='Covariance', cat='Stat_normalisedXsection' )
# create_covariance_matrix( norm_covariance_matrix, table_outfile)
# table_outfile=covariance_output_tempalte.format( path_to_DF=path_to_DF, channel = channel, label='Correlation', cat='Stat_normalisedXsection' )
# create_covariance_matrix( norm_correlation_matrix, table_outfile )
del unfolding
return hist_to_value_error_tuplelist( h_data_rebinned ), hist_to_value_error_tuplelist( h_unfolded_data ), hist_to_value_error_tuplelist( h_data_no_fakes ), covariance_matrix
def unfold_results(results, category, channel, h_truth, h_measured, h_response,
method):
global variable, path_to_JSON
h_data = value_error_tuplelist_to_hist(results, bin_edges[variable])
unfolding = Unfolding(h_truth, h_measured, h_response, method=method)
#turning off the unfolding errors for systematic samples
if category != 'central':
unfoldCfg.Hreco = 0
h_unfolded_data = unfolding.unfold(h_data)
#export the D and SV distributions
SVD_path = path_to_JSON + '/' + variable + '/unfolding_objects/' + channel + '/kv_' + str(
unfoldCfg.SVD_k_value) + '/'
make_folder_if_not_exists(SVD_path)
if method == 'TSVDUnfold':
SVDdist = TFile(
SVD_path + method + '_SVDdistributions_' + category + '.root',
'recreate')
directory = SVDdist.mkdir('SVDdist')
directory.cd()
unfolding.unfoldObject.GetD().Write()
unfolding.unfoldObject.GetSV().Write()
# unfolding.unfoldObject.GetUnfoldCovMatrix(data_covariance_matrix(h_data), unfoldCfg.SVD_n_toy).Write()
SVDdist.Close()
else:
SVDdist = TFile(
SVD_path + method + '_SVDdistributions_Hreco' +
str(unfoldCfg.Hreco) + '_' + category + '.root', 'recreate')
directory = SVDdist.mkdir('SVDdist')
directory.cd()
unfolding.unfoldObject.Impl().GetD().Write()
unfolding.unfoldObject.Impl().GetSV().Write()
h_truth.Write()
h_measured.Write()
h_response.Write()
# unfolding.unfoldObject.Impl().GetUnfoldCovMatrix(data_covariance_matrix(h_data), unfoldCfg.SVD_n_toy).Write()
SVDdist.Close()
#export the whole unfolding object if it doesn't exist
if method == 'TSVDUnfold':
unfolding_object_file_name = SVD_path + method + '_unfoldingObject_' + category + '.root'
else:
unfolding_object_file_name = SVD_path + method + '_unfoldingObject_Hreco' + str(
unfoldCfg.Hreco) + '_' + category + '.root'
if not os.path.isfile(unfolding_object_file_name):
unfoldingObjectFile = TFile(unfolding_object_file_name, 'recreate')
directory = unfoldingObjectFile.mkdir('unfoldingObject')
directory.cd()
if method == 'TSVDUnfold':
unfolding.unfoldObject.Write()
else:
unfolding.unfoldObject.Impl().Write()
unfoldingObjectFile.Close()
del unfolding
return hist_to_value_error_tuplelist(h_unfolded_data)
def main():
args, input_values_sets, json_input_files = parse_options()
results = {}
clear_old_df('tables/taufinding/')
for input_values, json_file in zip( input_values_sets, json_input_files ):
# print '\nProcessing', json_file
# Initialise the TauFinding class
regularisation_settings = TauFinding( input_values )
# Set additional elemtents
regularisation_settings.taus_to_test = get_tau_list(args.n_ticks_in_log)
variable = regularisation_settings.variable
channel = regularisation_settings.channel
com = regularisation_settings.centre_of_mass_energy
if 'muon' not in channel : continue
print 'Variable = {0}, channel = {1}, sqrt(s) = {2}'.format(variable, channel, com)
if args.run_measured_as_data:
regularisation_settings.taus_to_test = [0]
regularisation_settings.h_data = regularisation_settings.h_measured
df_chi2 = get_chi2s_of_tau_range(regularisation_settings, args)
if args.perform_varied_measured_unfolding_test:
h_data = hist_to_value_error_tuplelist(regularisation_settings.h_data)
h_data_varied = [(return_rnd_Poisson(val),return_rnd_Poisson(err)) for val, err in h_data ]
h_data_varied = value_error_tuplelist_to_hist(h_data_varied, reco_bin_edges_vis[variable])
regularisation_settings.h_data = h_data_varied
df_chi2_smeared = get_chi2s_of_tau_range(regularisation_settings, args, unfold_test=True)
print df_chi2_smeared
# No point in trying to find best tau if it is given as 0...
sys.exit()
# Find the corresponding Chi2 and write to file
df_chi2 = get_chi2s_of_tau_range(regularisation_settings, args)
print df_chi2
# Have the dataframes now - albeit read to a file
# Read in each one corresponding to their channel
# Find the best tau and print to screen
for channel in ['electron', 'muon', 'combined']:
chi2_cut = 0.005
path = regularisation_settings.outpath+'tbl_'+channel+'_tauscan.txt'
df_chi2 = get_df_from_file(path)
if df_chi2 is None: continue
print '\n', "1 - P(Chi2|NDF)", '\n', df_chi2, '\n'
# cutoff to be changed to 0.001 when able to
best_taus = interpolate_tau(chi2_cut, df_chi2)
chi2_to_plots(df_chi2, regularisation_settings, chi2_cut, channel)
print_results_to_screen(best_taus, channel)
return
def unfold_results(results, category, channel, h_truth, h_measured, h_response, method):
global variable, path_to_JSON
h_data = value_error_tuplelist_to_hist(results, bin_edges[variable])
unfolding = Unfolding(h_truth, h_measured, h_response, method=method)
#turning off the unfolding errors for systematic samples
if category != 'central':
unfoldCfg.Hreco = 0
h_unfolded_data = unfolding.unfold(h_data)
#export the D and SV distributions
SVD_path = path_to_JSON + '/' + variable + '/unfolding_objects/' + channel + '/kv_' + str(unfoldCfg.SVD_k_value) + '/'
make_folder_if_not_exists(SVD_path)
if method == 'TSVDUnfold':
SVDdist = TFile(SVD_path + method + '_SVDdistributions_' + category + '.root', 'recreate')
directory = SVDdist.mkdir('SVDdist')
directory.cd()
unfolding.unfoldObject.GetD().Write()
unfolding.unfoldObject.GetSV().Write()
# unfolding.unfoldObject.GetUnfoldCovMatrix(data_covariance_matrix(h_data), unfoldCfg.SVD_n_toy).Write()
SVDdist.Close()
else:
SVDdist = TFile(SVD_path + method + '_SVDdistributions_Hreco' + str(unfoldCfg.Hreco) + '_' + category + '.root', 'recreate')
directory = SVDdist.mkdir('SVDdist')
directory.cd()
unfolding.unfoldObject.Impl().GetD().Write()
unfolding.unfoldObject.Impl().GetSV().Write()
h_truth.Write()
h_measured.Write()
h_response.Write()
# unfolding.unfoldObject.Impl().GetUnfoldCovMatrix(data_covariance_matrix(h_data), unfoldCfg.SVD_n_toy).Write()
SVDdist.Close()
#export the whole unfolding object if it doesn't exist
if method == 'TSVDUnfold':
unfolding_object_file_name = SVD_path + method + '_unfoldingObject_' + category + '.root'
else:
unfolding_object_file_name = SVD_path + method + '_unfoldingObject_Hreco' + str(unfoldCfg.Hreco) + '_' + category + '.root'
if not os.path.isfile(unfolding_object_file_name):
unfoldingObjectFile = TFile(unfolding_object_file_name, 'recreate')
directory = unfoldingObjectFile.mkdir('unfoldingObject')
directory.cd()
if method == 'TSVDUnfold':
unfolding.unfoldObject.Write()
else:
unfolding.unfoldObject.Impl().Write()
unfoldingObjectFile.Close()
del unfolding
return hist_to_value_error_tuplelist(h_unfolded_data)
def unfold_results( results, category, channel, tau_value, h_truth, h_measured, h_response, h_fakes, method, visiblePS ):
global variable, path_to_JSON, options
edges = reco_bin_edges_full[variable]
if visiblePS:
edges = reco_bin_edges_vis[variable]
h_data = value_error_tuplelist_to_hist( results, edges )
# Remove fakes before unfolding
h_data = removeFakes( h_measured, h_fakes, h_data )
unfolding = Unfolding( h_data, h_truth, h_measured, h_response, h_fakes, method = method, tau = tau_value )
# turning off the unfolding errors for systematic samples
if not category == 'central':
unfoldCfg.error_treatment = 0
else:
unfoldCfg.error_treatment = options.error_treatment
h_unfolded_data = unfolding.unfold()
print "h_response bin edges : ", h_response
print "h_unfolded_data bin edges : ", h_unfolded_data
del unfolding
return hist_to_value_error_tuplelist( h_unfolded_data ), hist_to_value_error_tuplelist( h_data )
def get_unfolded_normalisation(TTJet_fit_results, category, channel):
global variable, met_type, path_to_JSON
h_truth, h_measured, h_response = get_unfold_histogram_tuple(
file_for_unfolding, variable, channel, met_type)
MADGRAPH_results = hist_to_value_error_tuplelist(h_truth)
POWHEG_results = hist_to_value_error_tuplelist(
get_unfold_histogram_tuple(file_for_powheg, variable, channel,
met_type)[0])
MCATNLO_results = hist_to_value_error_tuplelist(
get_unfold_histogram_tuple(file_for_mcatnlo, variable, channel,
met_type)[0])
matchingdown_results = hist_to_value_error_tuplelist(
get_unfold_histogram_tuple(file_for_matchingdown, variable, channel,
met_type)[0])
matchingup_results = hist_to_value_error_tuplelist(
get_unfold_histogram_tuple(file_for_matchingup, variable, channel,
met_type)[0])
scaledown_results = hist_to_value_error_tuplelist(
get_unfold_histogram_tuple(file_for_scaledown, variable, channel,
met_type)[0])
scaleup_results = hist_to_value_error_tuplelist(
get_unfold_histogram_tuple(file_for_scaleup, variable, channel,
met_type)[0])
TTJet_fit_results_unfolded = unfold_results(
TTJet_fit_results, category, channel, h_truth, h_measured, h_response,
'RooUnfoldSvd'
# 'TSVDUnfold'
)
normalisation_unfolded = {
'TTJet_measured': TTJet_fit_results,
'TTJet_unfolded': TTJet_fit_results_unfolded,
'MADGRAPH': MADGRAPH_results,
#other generators
'POWHEG': POWHEG_results,
'MCATNLO': MCATNLO_results,
#systematics
'matchingdown': matchingdown_results,
'matchingup': matchingup_results,
'scaledown': scaledown_results,
'scaleup': scaleup_results
}
write_data_to_JSON(
normalisation_unfolded, path_to_JSON + '/' + variable +
'/xsection_measurement_results' + '/kv' + str(unfoldCfg.SVD_k_value) +
'/' + category + '/normalisation_' + channel + '_' + met_type + '.txt')
return normalisation_unfolded
def __background_subtraction(self, histograms):
'''
Subtracts the backgrounds from data to give amount of ttbar in data.
Also adds all backgrounds to normalisation output
'''
ttjet_hist = clean_control_region(
histograms,
subtract=['QCD', 'V+Jets', 'SingleTop']
)
self.normalisation['TTJet'] = hist_to_value_error_tuplelist(ttjet_hist)
self.normalisation['data'] = hist_to_value_error_tuplelist(histograms['data'])
self.normalisation['TTJet_MC'] = hist_to_value_error_tuplelist(histograms['TTBar'])
self.normalisation['SingleTop'] = hist_to_value_error_tuplelist(histograms['SingleTop'])
self.normalisation['V+Jets'] = hist_to_value_error_tuplelist(histograms['V+Jets'])
self.normalisation['QCD'] = hist_to_value_error_tuplelist(histograms['QCD'])
return
def get_unfolded_normalisation(TTJet_fit_results, category, channel):
global variable, met_type, path_to_JSON
h_truth, h_measured, h_response = get_unfold_histogram_tuple(file_for_unfolding, variable, channel, met_type)
MADGRAPH_results = hist_to_value_error_tuplelist(h_truth)
POWHEG_results = hist_to_value_error_tuplelist(get_unfold_histogram_tuple(file_for_powheg, variable, channel, met_type)[0])
MCATNLO_results = hist_to_value_error_tuplelist(get_unfold_histogram_tuple(file_for_mcatnlo, variable, channel, met_type)[0])
matchingdown_results = hist_to_value_error_tuplelist(get_unfold_histogram_tuple(file_for_matchingdown, variable, channel, met_type)[0])
matchingup_results = hist_to_value_error_tuplelist(get_unfold_histogram_tuple(file_for_matchingup, variable, channel, met_type)[0])
scaledown_results = hist_to_value_error_tuplelist(get_unfold_histogram_tuple(file_for_scaledown, variable, channel, met_type)[0])
scaleup_results = hist_to_value_error_tuplelist(get_unfold_histogram_tuple(file_for_scaleup, variable, channel, met_type)[0])
TTJet_fit_results_unfolded = unfold_results(TTJet_fit_results,
category,
channel,
h_truth,
h_measured,
h_response,
'RooUnfoldSvd'
# 'TSVDUnfold'
)
normalisation_unfolded = {
'TTJet_measured' : TTJet_fit_results,
'TTJet_unfolded' : TTJet_fit_results_unfolded,
'MADGRAPH': MADGRAPH_results,
#other generators
'POWHEG': POWHEG_results,
'MCATNLO': MCATNLO_results,
#systematics
'matchingdown': matchingdown_results,
'matchingup': matchingup_results,
'scaledown': scaledown_results,
'scaleup': scaleup_results
}
write_data_to_JSON(normalisation_unfolded, path_to_JSON + '/' + variable + '/xsection_measurement_results' + '/kv' + str(unfoldCfg.SVD_k_value) + '/' + category + '/normalisation_' + channel + '_' + met_type + '.txt')
return normalisation_unfolded
def __background_subtraction(self, histograms):
'''
Subtracts the backgrounds from data to give amount of ttbar in data.
Also adds all backgrounds to normalisation output
'''
ttjet_hist = clean_control_region(
histograms, subtract=['QCD', 'V+Jets', 'SingleTop'])
self.normalisation['TTJet'] = hist_to_value_error_tuplelist(ttjet_hist)
self.normalisation['data'] = hist_to_value_error_tuplelist(
histograms['data'])
self.normalisation['TTJet_MC'] = hist_to_value_error_tuplelist(
histograms['TTBar'])
self.normalisation['SingleTop'] = hist_to_value_error_tuplelist(
histograms['SingleTop'])
self.normalisation['V+Jets'] = hist_to_value_error_tuplelist(
histograms['V+Jets'])
self.normalisation['QCD'] = hist_to_value_error_tuplelist(
histograms['QCD'])
return
def get_unfolded_normalisation( TTJet_normalisation_results, category, channel, tau_value, visiblePS ):
global com, luminosity, ttbar_xsection, method, variable, path_to_DF
global unfolding_files
files_for_systematics = {
'TTJets_massdown' : unfolding_files['file_for_massdown'],
'TTJets_massup' : unfolding_files['file_for_massup'],
'TTJets_factorisationdown' : unfolding_files['file_for_factorisationdown'],
'TTJets_factorisationup' : unfolding_files['file_for_factorisationup'],
'TTJets_renormalisationdown' : unfolding_files['file_for_renormalisationdown'],
'TTJets_renormalisationup' : unfolding_files['file_for_renormalisationup'],
'TTJets_combineddown' : unfolding_files['file_for_combineddown'],
'TTJets_combinedup' : unfolding_files['file_for_combinedup'],
'TTJets_alphaSdown' : unfolding_files['file_for_alphaSdown'],
'TTJets_alphaSup' : unfolding_files['file_for_alphaSup'],
'TTJets_matchingdown' : unfolding_files['file_for_matchingdown'],
'TTJets_matchingup' : unfolding_files['file_for_matchingup'],
'TTJets_isrdown' : unfolding_files['file_for_isrdown'],
'TTJets_isrup' : unfolding_files['file_for_isrup'],
# 'TTJets_fsrdown' : unfolding_files['file_for_fsrdown'],
'TTJets_fsrup' : unfolding_files['file_for_fsrup'],
'TTJets_uedown' : unfolding_files['file_for_uedown'],
'TTJets_ueup' : unfolding_files['file_for_ueup'],
'TTJets_topPt' : unfolding_files['file_for_topPtSystematic'],
'JES_down' : unfolding_files['file_for_jesdown'],
'JES_up' : unfolding_files['file_for_jesup'],
'JER_down' : unfolding_files['file_for_jerdown'],
'JER_up' : unfolding_files['file_for_jerup'],
'BJet_up' : unfolding_files['file_for_bjetup'],
'BJet_down' : unfolding_files['file_for_bjetdown'],
'LightJet_up' : unfolding_files['file_for_lightjetup'],
'LightJet_down' : unfolding_files['file_for_lightjetdown'],
'TTJets_hadronisation' : unfolding_files['file_for_powheg_herwig'],
'ElectronEnUp' : unfolding_files['file_for_ElectronEnUp'],
'ElectronEnDown' : unfolding_files['file_for_ElectronEnDown'],
'MuonEnUp' : unfolding_files['file_for_MuonEnUp'],
'MuonEnDown' : unfolding_files['file_for_MuonEnDown'],
'TauEnUp' : unfolding_files['file_for_TauEnUp'],
'TauEnDown' : unfolding_files['file_for_TauEnDown'],
'UnclusteredEnUp' : unfolding_files['file_for_UnclusteredEnUp'],
'UnclusteredEnDown' : unfolding_files['file_for_UnclusteredEnDown'],
'Muon_up' : unfolding_files['file_for_LeptonUp'],
'Muon_down' : unfolding_files['file_for_LeptonDown'],
'Electron_up' : unfolding_files['file_for_LeptonUp'],
'Electron_down' : unfolding_files['file_for_LeptonDown'],
'PileUp_up' : unfolding_files['file_for_PUUp'],
'PileUp_down' : unfolding_files['file_for_PUDown'],
'Top_Pt_reweight' : unfolding_files['file_for_ptreweight'],
}
h_truth, h_measured, h_response, h_fakes = None, None, None, None
# Uncertainties by changing the response matrix
if category in files_for_systematics :
print 'Doing category',category,'by changing response matrix'
h_truth, h_measured, h_response, h_fakes = get_unfold_histogram_tuple(
inputfile = files_for_systematics[category],
variable = variable,
channel = channel,
centre_of_mass = com,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
# PDF Uncertainties
elif category in pdf_uncertainties:
print 'Doing category',category,'by changing response matrix'
h_truth, h_measured, h_response, h_fakes = get_unfold_histogram_tuple(
inputfile = unfolding_files['files_for_pdfs'][category],
variable = variable,
channel = channel,
centre_of_mass = com,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
# Central and systematics where you just change input MC
else:
h_truth, h_measured, h_response, h_fakes = get_unfold_histogram_tuple(
inputfile = unfolding_files['file_for_unfolding'],
variable = variable,
channel = channel,
centre_of_mass = com,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
# Unfold current normalisation measurements
TTJet_normalisation_results, TTJet_normalisation_results_unfolded, TTJet_normalisation_results_withoutFakes, covariance_matrix = unfold_results(
TTJet_normalisation_results,
category,
channel,
tau_value,
h_truth,
h_measured,
h_response,
h_fakes,
method,
visiblePS,
)
# Store TTJet yields after background subtraction, after background subtraction without fakes and after Unfolding
normalisation_unfolded = {
'TTJet_measured' : TTJet_normalisation_results,
'TTJet_measured_withoutFakes' : TTJet_normalisation_results_withoutFakes,
'TTJet_unfolded' : TTJet_normalisation_results_unfolded,
}
# Return truth of different generators for comparison to data in 04
if category == 'central':
h_truth_massdown, _, _, _ = get_unfold_histogram_tuple(
inputfile = unfolding_files['file_for_massdown'],
variable = variable,
channel = channel,
centre_of_mass = com,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_massup, _, _, _ = get_unfold_histogram_tuple(
inputfile = unfolding_files['file_for_massup'],
variable = variable,
channel = channel,
centre_of_mass = com,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
# h_truth_fsrdown, _, _, _ = get_unfold_histogram_tuple(
# inputfile = unfolding_files['file_for_fsrdown'],
# variable = variable,
# channel = channel,
# centre_of_mass = com,
# ttbar_xsection = ttbar_xsection,
# luminosity = luminosity,
# load_fakes = True,
# visiblePS = visiblePS,
# )
h_truth_fsrup, _, _, _ = get_unfold_histogram_tuple(
inputfile = unfolding_files['file_for_fsrup'],
variable = variable,
channel = channel,
centre_of_mass = com,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_isrdown, _, _, _ = get_unfold_histogram_tuple(
inputfile = unfolding_files['file_for_isrdown'],
variable = variable,
channel = channel,
centre_of_mass = com,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_isrup, _, _, _ = get_unfold_histogram_tuple(
inputfile = unfolding_files['file_for_isrup'],
variable = variable,
channel = channel,
centre_of_mass = com,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_uedown, _, _, _ = get_unfold_histogram_tuple(
inputfile = unfolding_files['file_for_uedown'],
variable = variable,
channel = channel,
centre_of_mass = com,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_ueup, _, _, _ = get_unfold_histogram_tuple(
inputfile = unfolding_files['file_for_ueup'],
variable = variable,
channel = channel,
centre_of_mass = com,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_powhegPythia8, _, _, _ = get_unfold_histogram_tuple(
inputfile = unfolding_files['file_for_powhegPythia8'],
variable = variable,
channel = channel,
centre_of_mass = com,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
# h_truth_amcatnlo, _, _, _ = get_unfold_histogram_tuple(
# inputfile = unfolding_files['file_for_amcatnlo'],
# variable = variable,
# channel = channel,
# centre_of_mass = com,
# ttbar_xsection = ttbar_xsection,
# luminosity = luminosity,
# load_fakes = True,
# visiblePS = visiblePS,
# )
# h_truth_madgraphMLM, _, _, _ = get_unfold_histogram_tuple(
# inputfile = unfolding_files['file_for_madgraphMLM'],
# variable = variable,
# channel = channel,
# centre_of_mass = com,
# ttbar_xsection = ttbar_xsection,
# luminosity = luminosity,
# load_fakes = True,
# visiblePS = visiblePS,
# )
h_truth_powheg_herwig, _, _, _ = get_unfold_histogram_tuple(
inputfile = unfolding_files['file_for_powheg_herwig'],
variable = variable,
channel = channel,
centre_of_mass = com,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
normalisation_unfolded['powhegPythia8'] = hist_to_value_error_tuplelist( h_truth_powhegPythia8 )
# normalisation_unfolded['amcatnlo'] = hist_to_value_error_tuplelist( h_truth_madgraphMLM )
# normalisation_unfolded['madgraphMLM'] = hist_to_value_error_tuplelist( h_truth_amcatnlo )
normalisation_unfolded['powhegHerwig'] = hist_to_value_error_tuplelist( h_truth_powheg_herwig )
normalisation_unfolded['massdown'] = hist_to_value_error_tuplelist( h_truth_massdown )
normalisation_unfolded['massup'] = hist_to_value_error_tuplelist( h_truth_massup )
normalisation_unfolded['isrdown'] = hist_to_value_error_tuplelist( h_truth_isrdown )
normalisation_unfolded['isrup'] = hist_to_value_error_tuplelist( h_truth_isrup )
# normalisation_unfolded['fsrdown'] = hist_to_value_error_tuplelist( h_truth_fsrdown )
normalisation_unfolded['fsrup'] = hist_to_value_error_tuplelist( h_truth_fsrup )
normalisation_unfolded['uedown'] = hist_to_value_error_tuplelist( h_truth_uedown )
normalisation_unfolded['ueup'] = hist_to_value_error_tuplelist( h_truth_ueup )
# Write all normalisations in unfolded binning scheme to dataframes
file_template = '{path_to_DF}/{category}/unfolded_normalisation_{channel}_{method}.txt'
write_02(normalisation_unfolded, file_template, path_to_DF, category, channel, method)
return normalisation_unfolded, covariance_matrix
def check_multiple_data_multiple_unfolding(
input_file, method, channel, variable,
responseMatrix,
n_toy_data, output_folder,
tau_value=-1
):
'''
Loops through a n_toy_data of pseudo data,
unfolds the pseudo data and compares it to the MC truth
'''
# same unfolding input, different data
get_folder = input_file.Get
pulls = []
add_pull = pulls.append
histograms = []
add_histograms = histograms.append
print('Reading toy MC')
start1 = time()
data_range = range(0, n_toy_data)
for nth_toy_data in range(0, n_toy_data + 1): # read all of them (easier)
if nth_toy_data in data_range:
tpl = '{channel}/{variable}/toy_{nth}'
folder_mc = tpl.format(channel=channel, variable=variable,
nth=nth_toy_data+1)
folder_mc = get_folder(folder_mc)
add_histograms(get_measured_histogram(folder_mc))
else:
add_histograms(0)
print('Done reading toy MC in', time() - start1, 's')
# Get truth and measured histograms
h_truth = get_truth_histogram( get_folder('{channel}/{variable}/Original'.format(channel=channel, variable=variable) ) )
h_measured = get_measured_histogram( get_folder('{channel}/{variable}/Original'.format(channel=channel, variable=variable) ) )
# Set response matrix
h_response = responseMatrix
# Make sure the pseudo data to be unfolded has the same integral as the response matrix
measured_from_response = asrootpy( h_response.ProjectionX('px',1) )
truth_from_response = asrootpy( h_response.ProjectionY() )
truthScale = truth_from_response.Integral() / h_truth.Integral()
h_truth.Scale( truthScale )
h_measured.Scale( truthScale )
for nth_toy_data in data_range:
if nth_toy_data % 100 == 0 :
print(
'Doing data no', nth_toy_data)
h_data = histograms[nth_toy_data]
h_data.Scale( truthScale )
unfolding_obj = Unfolding(
h_data,
h_truth, h_data, h_response, method=method, k_value=-1,
tau=tau_value)
unfold, get_pull = unfolding_obj.unfold, unfolding_obj.pull
reset = unfolding_obj.Reset
unfold()
pull = get_pull()
diff = unfolding_obj.unfolded_data - h_truth
diff_tuple = hist_to_value_error_tuplelist(diff)
truth_tuple = hist_to_value_error_tuplelist(unfolding_obj.truth)
bias = []
sumBias2 = 0
for d, t in zip(diff_tuple, truth_tuple):
b = d[0] / t[0]
bias.append(b)
unfolded = unfolding_obj.unfolded_data
unfolded_tuple = hist_to_value_error_tuplelist(unfolded)
all_data = {'unfolded': unfolded_tuple,
'difference': diff_tuple,
'truth': truth_tuple,
'bias':bias,
'pull': pull,
'nth_toy_data': nth_toy_data
}
add_pull(all_data)
reset()
output_file_name = save_pulls(pulls, method,
channel, tau_value, output_folder)
return output_file_name
def background_subtraction(self, histograms):
ttjet_hist = clean_control_region(
histograms, subtract=['QCD', 'V+Jets', 'SingleTop'])
self.normalisation['TTJet'] = hist_to_value_error_tuplelist(ttjet_hist)
def check_multiple_data_multiple_unfolding(input_file,
method,
channel,
variable,
responseMatrix,
n_toy_data,
output_folder,
tau_value=-1):
'''
Loops through a n_toy_data of pseudo data,
unfolds the pseudo data and compares it to the MC truth
'''
# same unfolding input, different data
get_folder = input_file.Get
pulls = []
add_pull = pulls.append
histograms = []
add_histograms = histograms.append
truth_histograms = []
dirs = None
for path, dir, objects in input_file.walk(maxdepth=0):
dirs = dir
for dir in dirs:
print('Reading toy MC')
start1 = time()
data_range = range(0, n_toy_data)
for nth_toy_data in range(0,
n_toy_data + 1): # read all of them (easier)
if nth_toy_data in data_range:
tpl = '{dir}/{channel}/{variable}/toy_{nth}'
folder_mc = tpl.format(dir=dir,
channel=channel,
variable=variable,
nth=nth_toy_data + 1)
folder_mc = get_folder(folder_mc)
add_histograms(get_measured_histogram(folder_mc))
truth_histograms.append(get_truth_histogram(folder_mc))
else:
add_histograms(0)
print('Done reading toy MC in', time() - start1, 's')
# Get truth and measured histograms
h_truth = get_truth_histogram(
get_folder('{dir}/{channel}/{variable}/Original'.format(
dir=dir, channel=channel, variable=variable)))
h_measured = get_measured_histogram(
get_folder('{dir}/{channel}/{variable}/Original'.format(
dir=dir, channel=channel, variable=variable)))
# Set response matrix
h_response = generate_toy_MC_from_2Ddistribution(responseMatrix)
# Make sure the response matrix has the same normalisatio as the pseudo data to be unfolded
truthScale = h_truth.integral(overflow=True) / h_response.integral(
overflow=True)
h_response.Scale(truthScale)
# measured_from_response = asrootpy( h_response.ProjectionX('px',1) )
# truth_from_response = asrootpy( h_response.ProjectionY() )
for nth_toy_data in data_range:
if nth_toy_data % 100 == 0:
print('Doing data no', nth_toy_data)
h_data = histograms[nth_toy_data]
# h_truth = truth_histograms[nth_toy_data]
unfolding_obj = Unfolding(h_data,
h_truth,
h_data,
h_response,
method=method,
tau=tau_value)
unfold, get_pull = unfolding_obj.unfold, unfolding_obj.pull
reset = unfolding_obj.Reset
unfold()
# print ('Measured :',list(h_data.y()))
# print ('Unfolded :',list( unfolding_obj.unfolded_data.y() ))
pull = get_pull()
# print ('Pull :',pull)
diff = unfolding_obj.unfolded_data - h_truth
# print ('Diff :',list(diff.y()))
diff_tuple = hist_to_value_error_tuplelist(diff)
truth_tuple = hist_to_value_error_tuplelist(unfolding_obj.truth)
bias = []
sumBias2 = 0
for d, t in zip(diff_tuple, truth_tuple):
b = d[0] / t[0]
bias.append(b)
unfolded = unfolding_obj.unfolded_data
unfolded_tuple = hist_to_value_error_tuplelist(unfolded)
all_data = {
'unfolded': unfolded_tuple,
'difference': diff_tuple,
'truth': truth_tuple,
'bias': bias,
'pull': pull,
'nth_toy_data': nth_toy_data
}
add_pull(all_data)
reset()
output_file_name = save_pulls(pulls, method, channel, tau_value,
output_folder)
return output_file_name
def background_subtraction(self, histograms):
ttjet_hist = clean_control_region(histograms,
subtract=['QCD', 'V+Jets', 'SingleTop'])
self.normalisation[
'TTJet'] = hist_to_value_error_tuplelist(ttjet_hist)
def get_unfolded_normalisation( TTJet_fit_results, category, channel, tau_value, visiblePS ):
global centre_of_mass, luminosity, ttbar_xsection, method
global variable, met_type, path_to_JSON, file_for_unfolding, file_for_powheg_pythia, file_for_powheg_herwig, file_for_ptreweight, files_for_pdfs
global file_for_powhegPythia8, file_for_madgraphMLM, file_for_amcatnlo, file_for_amcatnlo_herwig
# global file_for_matchingdown, file_for_matchingup
global file_for_scaledown, file_for_scaleup
global file_for_massdown, file_for_massup
global ttbar_generator_systematics, ttbar_theory_systematics, pdf_uncertainties
global use_ptreweight
files_for_systematics = {
ttbar_theory_systematic_prefix + 'scaledown' : file_for_scaledown,
ttbar_theory_systematic_prefix + 'scaleup' : file_for_scaleup,
ttbar_theory_systematic_prefix + 'massdown' : file_for_massdown,
ttbar_theory_systematic_prefix + 'massup' : file_for_massup,
ttbar_theory_systematic_prefix + 'factorisationdown' : file_for_factorisationdown,
ttbar_theory_systematic_prefix + 'factorisationup' : file_for_factorisationup,
ttbar_theory_systematic_prefix + 'renormalisationdown' : file_for_renormalisationdown,
ttbar_theory_systematic_prefix + 'renormalisationup' : file_for_renormalisationup,
ttbar_theory_systematic_prefix + 'combineddown' : file_for_combineddown,
ttbar_theory_systematic_prefix + 'combinedup' : file_for_combinedup,
ttbar_theory_systematic_prefix + 'alphaSdown' : file_for_alphaSdown,
ttbar_theory_systematic_prefix + 'alphaSup' : file_for_alphaSup,
'JES_down' : file_for_jesdown,
'JES_up' : file_for_jesup,
'JER_down' : file_for_jerdown,
'JER_up' : file_for_jerup,
'BJet_up' : file_for_bjetup,
'BJet_down' : file_for_bjetdown,
'LightJet_up' : file_for_lightjetup,
'LightJet_down' : file_for_lightjetdown,
ttbar_theory_systematic_prefix + 'hadronisation' : file_for_powheg_herwig,
ttbar_theory_systematic_prefix + 'NLOgenerator' : file_for_amcatnlo,
'ElectronEnUp' : file_for_ElectronEnUp,
'ElectronEnDown' : file_for_ElectronEnDown,
'MuonEnUp' : file_for_MuonEnUp,
'MuonEnDown' : file_for_MuonEnDown,
'TauEnUp' : file_for_TauEnUp,
'TauEnDown' : file_for_TauEnDown,
'UnclusteredEnUp' : file_for_UnclusteredEnUp,
'UnclusteredEnDown' : file_for_UnclusteredEnDown,
'Muon_up' : file_for_LeptonUp,
'Muon_down' : file_for_LeptonDown,
'Electron_up' : file_for_LeptonUp,
'Electron_down' : file_for_LeptonDown,
'PileUp_up' : file_for_PUUp,
'PileUp_down' : file_for_PUDown,
}
h_truth, h_measured, h_response, h_fakes = None, None, None, None
# Systematics where you change the response matrix
if category in files_for_systematics :
print 'Doing category',category,'by changing response matrix'
h_truth, h_measured, h_response, h_fakes = get_unfold_histogram_tuple( inputfile = files_for_systematics[category],
variable = variable,
channel = channel,
met_type = met_type,
centre_of_mass = centre_of_mass,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
elif category in pdf_uncertainties:
h_truth, h_measured, h_response, h_fakes = get_unfold_histogram_tuple( inputfile = files_for_pdfs[category],
variable = variable,
channel = channel,
met_type = met_type,
centre_of_mass = centre_of_mass,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
# Central and systematics where you just change input MC
else:
h_truth, h_measured, h_response, h_fakes = get_unfold_histogram_tuple( inputfile = file_for_unfolding,
variable = variable,
channel = channel,
met_type = met_type,
centre_of_mass = centre_of_mass,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
# central_results = hist_to_value_error_tuplelist( h_truth )
TTJet_fit_results_unfolded, TTJet_fit_results_withoutFakes = unfold_results( TTJet_fit_results,
category,
channel,
tau_value,
h_truth,
h_measured,
h_response,
h_fakes,
method,
visiblePS,
)
normalisation_unfolded = {
'TTJet_measured' : TTJet_fit_results,
'TTJet_measured_withoutFakes' : TTJet_fit_results_withoutFakes,
'TTJet_unfolded' : TTJet_fit_results_unfolded
}
#
# THESE ARE FOR GETTING THE HISTOGRAMS FOR COMPARING WITH UNFOLDED DATA
#
if category == 'central':
h_truth_scaledown, _, _, _ = get_unfold_histogram_tuple( inputfile = file_for_scaledown,
variable = variable,
channel = channel,
met_type = met_type,
centre_of_mass = centre_of_mass,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_scaleup, _, _, _ = get_unfold_histogram_tuple( inputfile = file_for_scaleup,
variable = variable,
channel = channel,
met_type = met_type,
centre_of_mass = centre_of_mass,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_massdown, _, _, _ = get_unfold_histogram_tuple( inputfile = file_for_massdown,
variable = variable,
channel = channel,
met_type = met_type,
centre_of_mass = centre_of_mass,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_massup, _, _, _ = get_unfold_histogram_tuple( inputfile = file_for_massup,
variable = variable,
channel = channel,
met_type = met_type,
centre_of_mass = centre_of_mass,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_powhegPythia8, _, _, _ = get_unfold_histogram_tuple( inputfile = file_for_powhegPythia8,
variable = variable,
channel = channel,
met_type = met_type,
centre_of_mass = centre_of_mass,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_amcatnlo, _, _, _ = get_unfold_histogram_tuple( inputfile = file_for_amcatnlo,
variable = variable,
channel = channel,
met_type = met_type,
centre_of_mass = centre_of_mass,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_madgraphMLM, _, _, _ = get_unfold_histogram_tuple( inputfile = file_for_madgraphMLM,
variable = variable,
channel = channel,
met_type = met_type,
centre_of_mass = centre_of_mass,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
h_truth_powheg_herwig, _, _, _ = get_unfold_histogram_tuple( inputfile = file_for_powheg_herwig,
variable = variable,
channel = channel,
met_type = met_type,
centre_of_mass = centre_of_mass,
ttbar_xsection = ttbar_xsection,
luminosity = luminosity,
load_fakes = True,
visiblePS = visiblePS,
)
# h_truth_amcatnlo_herwig, _, _, _ = get_unfold_histogram_tuple( inputfile = file_for_amcatnlo_herwig,
# variable = variable,
# channel = channel,
# met_type = met_type,
# centre_of_mass = centre_of_mass,
# ttbar_xsection = ttbar_xsection,
# luminosity = luminosity,
# load_fakes = True,
# visiblePS = visiblePS,
# )
# MADGRAPH_ptreweight_results = hist_to_value_error_tuplelist( h_truth_ptreweight )
# POWHEG_PYTHIA_results = hist_to_value_error_tuplelist( h_truth_POWHEG_PYTHIA )
# MCATNLO_results = None
powhegPythia8_results = hist_to_value_error_tuplelist( h_truth_powhegPythia8 )
madgraphMLM_results = hist_to_value_error_tuplelist( h_truth_madgraphMLM )
amcatnloPythia8_results = hist_to_value_error_tuplelist( h_truth_amcatnlo )
powheg_herwig_results = hist_to_value_error_tuplelist( h_truth_powheg_herwig )
# amcatnlo_herwig_results = hist_to_value_error_tuplelist( h_truth_amcatnlo_herwig )
# matchingdown_results = hist_to_value_error_tuplelist( h_truth_matchingdown )
# matchingup_results = hist_to_value_error_tuplelist( h_truth_matchingup )
scaledown_results = hist_to_value_error_tuplelist( h_truth_scaledown )
scaleup_results = hist_to_value_error_tuplelist( h_truth_scaleup )
massdown_results = hist_to_value_error_tuplelist( h_truth_massdown )
massup_results = hist_to_value_error_tuplelist( h_truth_massup )
normalisation_unfolded['powhegPythia8'] = powhegPythia8_results
normalisation_unfolded['amcatnlo'] = amcatnloPythia8_results
normalisation_unfolded['madgraphMLM'] = madgraphMLM_results
normalisation_unfolded['powhegHerwig'] = powheg_herwig_results
# normalisation_unfolded['amcatnloHerwig'] = amcatnlo_herwig_results
normalisation_unfolded['scaledown'] = scaledown_results
normalisation_unfolded['scaleup'] = scaleup_results
normalisation_unfolded['massdown'] = massdown_results
normalisation_unfolded['massup'] = massup_results
return normalisation_unfolded
'/storage/TopQuarkGroup/mc/8TeV/NoSkimUnfolding/v10/TTJets_scaledown_TuneZ2star_8TeV-madgraph-tauola/unfolding_v10_Summer12_DR53X-PU_S10_START53_V7A-v1_NoSkim/TTJets-scaledown_nTuple_53X_mc_merged_001.root'
)
test_file = root_open('test_unfolded.root')
test1 = test_file.Get(channel + '_MET__TTJet__TTJetsMatching__plus')
test2 = test_file.Get(channel + '_MET__TTJet__TTJetsMatching__minus')
test3 = test_file.Get(channel + '_MET__TTJet__TTJetsScale__plus')
test4 = test_file.Get(channel + '_MET__TTJet__TTJetsScale__minus')
test1.Sumw2()
test2.Sumw2()
test3.Sumw2()
test4.Sumw2()
folder = 'unfolding_MET_analyser_' + channel + '_channel_patMETsPFlow'
ref1 = hist_to_value_error_tuplelist(
unfolding_file1.Get(folder + '/truth_AsymBins'))
ref2 = hist_to_value_error_tuplelist(
unfolding_file2.Get(folder + '/truth_AsymBins'))
ref3 = hist_to_value_error_tuplelist(
unfolding_file3.Get(folder + '/truth_AsymBins'))
ref4 = hist_to_value_error_tuplelist(
unfolding_file4.Get(folder + '/truth_AsymBins'))
ref1 = value_error_tuplelist_to_hist(ref1, bin_edges['MET'])
ref2 = value_error_tuplelist_to_hist(ref2, bin_edges['MET'])
ref3 = value_error_tuplelist_to_hist(ref3, bin_edges['MET'])
ref4 = value_error_tuplelist_to_hist(ref4, bin_edges['MET'])
normalise([test1, test2, test3, test4, ref1, ref2, ref3, ref4])
draw_pair(test1, ref1, 'matching_up')
draw_pair(test2, ref2, 'matching_down')
def calculate_xsection( nEventsHistogram, variable ):
resultsAsTuple = hist_to_value_error_tuplelist( nEventsHistogram )
normalised_xsection, _, _, _ = calculate_normalised_xsection( resultsAsTuple, bin_widths_visiblePS[variable], False )
return value_error_tuplelist_to_hist(normalised_xsection, bin_edges_vis[variable])
def main():
args, input_values_sets, json_input_files = parse_options()
results = {}
clear_old_df('tables/taufinding/')
for input_values, json_file in zip( input_values_sets, json_input_files ):
if 'combined' in json_file: continue
# Initialise the TauFinding class
regularisation_settings = TauFinding( input_values )
variable = regularisation_settings.variable
channel = regularisation_settings.channel
com = regularisation_settings.centre_of_mass_energy
# Specific channel or variable
if args.ch:
if args.ch not in channel: continue
if args.var:
if args.var not in variable: continue
print 'Running for:'
print 'Variable = {0}, channel = {1}, sqrt(s) = {2}'.format(variable, channel, com)
# Set additional elements
regularisation_settings.taus_to_test = get_tau_values(args.n_tau_in_log)
isTauCalculator = True
# Specific unfolding tests go here
if args.specific_tau is not None:
regularisation_settings.taus_to_test = [args.specific_tau]
df_chi2_specific_tau = get_chi2(regularisation_settings, args)
isTauCalculator = False
if args.run_measured_as_data:
regularisation_settings.taus_to_test = [0]
regularisation_settings.h_data = regularisation_settings.h_measured
df_chi2_measured = get_chi2(regularisation_settings, args)
isTauCalculator = False
if args.run_smeared_measured_as_data:
regularisation_settings.taus_to_test = [0]
regularisation_settings.h_data = regularisation_settings.h_measured
h_data = hist_to_value_error_tuplelist(regularisation_settings.h_data)
h_data_varied = [(return_rnd_Poisson(val),return_rnd_Poisson(err)) for val, err in h_data ]
h_data_varied = value_error_tuplelist_to_hist(h_data_varied, reco_bin_edges_vis[variable])
regularisation_settings.h_data = h_data_varied
df_chi2_smeared = get_chi2(regularisation_settings, args, smearing_test=True)
isTauCalculator = False
# Dont need to calculate chi2 for given tau tests
if not isTauCalculator: continue
# Find Chi2 for each tau and write to file
df_chi2 = get_chi2(regularisation_settings, args)
# Dont need to calculate tau for given tests
if not isTauCalculator: sys.exit()
# Have the dataframes now - albeit read to a file
# Read in each one corresponding to their channel
# Find the best tau and print to screen
for channel in ['electron', 'muon', 'combined']:
chi2_cut = 0.005
path = regularisation_settings.outpath+'tbl_'+channel+'_tauscan.txt'
df_chi2 = get_df_from_file(path)
if df_chi2 is None: continue
print '\n', "1 - P(Chi2|NDF)", '\n', df_chi2, '\n'
# cutoff to be changed to 0.001 when able to
best_taus = interpolate_tau(chi2_cut, df_chi2)
chi2_to_plots(args, df_chi2, regularisation_settings, chi2_cut, channel)
print_results_to_screen(best_taus, channel)
return
def calculate_xsection( nEventsHistogram, variable ):
resultsAsTuple = hist_to_value_error_tuplelist( nEventsHistogram )
normalised_xsection = calculate_normalised_xsection( resultsAsTuple, bin_widths_visiblePS[variable], False )
return value_error_tuplelist_to_hist(normalised_xsection, bin_edges_vis[variable])
unfolding_file3 = root_open('/storage/TopQuarkGroup/mc/8TeV/NoSkimUnfolding/v10/TTJets_scaleup_TuneZ2star_8TeV-madgraph-tauola/unfolding_v10_Summer12_DR53X-PU_S10_START53_V7A-v1_NoSkim/TTJets-scaleup_nTuple_53X_mc_merged_001.root')
unfolding_file4 = root_open('/storage/TopQuarkGroup/mc/8TeV/NoSkimUnfolding/v10/TTJets_scaledown_TuneZ2star_8TeV-madgraph-tauola/unfolding_v10_Summer12_DR53X-PU_S10_START53_V7A-v1_NoSkim/TTJets-scaledown_nTuple_53X_mc_merged_001.root')
test_file = root_open('test_unfolded.root')
test1 = test_file.Get(channel + '_MET__TTJet__TTJetsMatching__plus')
test2 = test_file.Get(channel + '_MET__TTJet__TTJetsMatching__minus')
test3 = test_file.Get(channel + '_MET__TTJet__TTJetsScale__plus')
test4 = test_file.Get(channel + '_MET__TTJet__TTJetsScale__minus')
test1.Sumw2()
test2.Sumw2()
test3.Sumw2()
test4.Sumw2()
folder = 'unfolding_MET_analyser_' + channel + '_channel_patMETsPFlow'
ref1 = hist_to_value_error_tuplelist(unfolding_file1.Get(folder + '/truth_AsymBins'))
ref2 = hist_to_value_error_tuplelist(unfolding_file2.Get(folder + '/truth_AsymBins'))
ref3 = hist_to_value_error_tuplelist(unfolding_file3.Get(folder + '/truth_AsymBins'))
ref4 = hist_to_value_error_tuplelist(unfolding_file4.Get(folder + '/truth_AsymBins'))
ref1 = value_error_tuplelist_to_hist(ref1, bin_edges['MET'])
ref2 = value_error_tuplelist_to_hist(ref2, bin_edges['MET'])
ref3 = value_error_tuplelist_to_hist(ref3, bin_edges['MET'])
ref4 = value_error_tuplelist_to_hist(ref4, bin_edges['MET'])
normalise([test1, test2, test3, test4, ref1, ref2, ref3, ref4])
draw_pair(test1, ref1, 'matching_up')
draw_pair(test2, ref2, 'matching_down')
draw_pair(test3, ref3, 'scale_up')
draw_pair(test4, ref4, 'scale_down')
