def compare_combine_before_after_unfolding(measurement='normalised_xsection',
add_before_unfolding=False):
file_template = 'data/normalisation/background_subtraction/13TeV/'
file_template += '{variable}/VisiblePS/central/'
file_template += '{measurement}_{channel}_RooUnfold{method}.txt'
variables = [
'MET', 'HT', 'ST', 'NJets', 'lepton_pt', 'abs_lepton_eta', 'WPT'
]
for variable in variables:
combineBefore = file_template.format(variable=variable,
method='Svd',
channel='combinedBeforeUnfolding',
measurement=measurement)
combineAfter = file_template.format(variable=variable,
method='Svd',
channel='combined',
measurement=measurement)
data = read_data_from_JSON(combineBefore)
before_unfolding = data['TTJet_measured']
combineBefore_data = data['TTJet_unfolded']
combineAfter_data = read_data_from_JSON(combineAfter)['TTJet_unfolded']
h_combineBefore = value_error_tuplelist_to_hist(
combineBefore_data, bin_edges_vis[variable])
h_combineAfter = value_error_tuplelist_to_hist(combineAfter_data,
bin_edges_vis[variable])
h_before_unfolding = value_error_tuplelist_to_hist(
before_unfolding, bin_edges_vis[variable])
properties = Histogram_properties()
properties.name = '{0}_compare_combine_before_after_unfolding_{1}'.format(
measurement, variable)
properties.title = 'Comparison of combining before/after unfolding'
properties.path = 'plots'
properties.has_ratio = True
properties.xerr = True
properties.x_limits = (bin_edges_vis[variable][0],
bin_edges_vis[variable][-1])
properties.x_axis_title = variables_latex[variable]
if 'xsection' in measurement:
properties.y_axis_title = r'$\frac{1}{\sigma} \frac{d\sigma}{d' + \
variables_latex[variable] + '}$'
else:
properties.y_axis_title = r'$t\bar{t}$ normalisation'
histograms = {
'Combine before unfolding': h_combineBefore,
'Combine after unfolding': h_combineAfter
}
if add_before_unfolding:
histograms['before unfolding'] = h_before_unfolding
properties.name += '_ext'
properties.has_ratio = False
plot = Plot(histograms, properties)
plot.draw_method = 'errorbar'
compare_histograms(plot)
def compare_unfolding_methods(measurement='normalised_xsection',
add_before_unfolding=False,
channel='combined'):
file_template = '/hdfs/TopQuarkGroup/run2/dpsData/'
file_template += 'data/normalisation/background_subtraction/13TeV/'
file_template += '{variable}/VisiblePS/central/'
file_template += '{measurement}_{channel}_RooUnfold{method}.txt'
variables = [
'MET', 'HT', 'ST', 'NJets', 'lepton_pt', 'abs_lepton_eta', 'WPT'
]
for variable in variables:
svd = file_template.format(variable=variable,
method='Svd',
channel=channel,
measurement=measurement)
bayes = file_template.format(variable=variable,
method='Bayes',
channel=channel,
measurement=measurement)
data = read_data_from_JSON(svd)
before_unfolding = data['TTJet_measured_withoutFakes']
svd_data = data['TTJet_unfolded']
bayes_data = read_data_from_JSON(bayes)['TTJet_unfolded']
h_svd = value_error_tuplelist_to_hist(svd_data,
bin_edges_vis[variable])
h_bayes = value_error_tuplelist_to_hist(bayes_data,
bin_edges_vis[variable])
h_before_unfolding = value_error_tuplelist_to_hist(
before_unfolding, bin_edges_vis[variable])
properties = Histogram_properties()
properties.name = '{0}_compare_unfolding_methods_{1}_{2}'.format(
measurement, variable, channel)
properties.title = 'Comparison of unfolding methods'
properties.path = 'plots'
properties.has_ratio = True
properties.xerr = True
properties.x_limits = (bin_edges_vis[variable][0],
bin_edges_vis[variable][-1])
properties.x_axis_title = variables_latex[variable]
if 'xsection' in measurement:
properties.y_axis_title = r'$\frac{1}{\sigma} \frac{d\sigma}{d' + \
variables_latex[variable] + '}$'
else:
properties.y_axis_title = r'$t\bar{t}$ normalisation'
histograms = {'SVD': h_svd, 'Bayes': h_bayes}
if add_before_unfolding:
histograms['before unfolding'] = h_before_unfolding
properties.name += '_ext'
properties.has_ratio = False
plot = Plot(histograms, properties)
plot.draw_method = 'errorbar'
compare_histograms(plot)
def compare_combine_before_after_unfolding(measurement='normalised_xsection',
add_before_unfolding=False):
file_template = 'data/normalisation/background_subtraction/13TeV/'
file_template += '{variable}/VisiblePS/central/'
file_template += '{measurement}_{channel}_RooUnfold{method}.txt'
variables = ['MET', 'HT', 'ST', 'NJets',
'lepton_pt', 'abs_lepton_eta', 'WPT']
for variable in variables:
combineBefore = file_template.format(
variable=variable,
method='Svd',
channel='combinedBeforeUnfolding',
measurement=measurement)
combineAfter = file_template.format(
variable=variable,
method='Svd',
channel='combined',
measurement=measurement)
data = read_data_from_JSON(combineBefore)
before_unfolding = data['TTJet_measured']
combineBefore_data = data['TTJet_unfolded']
combineAfter_data = read_data_from_JSON(combineAfter)['TTJet_unfolded']
h_combineBefore = value_error_tuplelist_to_hist(
combineBefore_data, bin_edges_vis[variable])
h_combineAfter = value_error_tuplelist_to_hist(
combineAfter_data, bin_edges_vis[variable])
h_before_unfolding = value_error_tuplelist_to_hist(
before_unfolding, bin_edges_vis[variable])
properties = Histogram_properties()
properties.name = '{0}_compare_combine_before_after_unfolding_{1}'.format(
measurement, variable)
properties.title = 'Comparison of combining before/after unfolding'
properties.path = 'plots'
properties.has_ratio = True
properties.xerr = True
properties.x_limits = (
bin_edges_vis[variable][0], bin_edges_vis[variable][-1])
properties.x_axis_title = variables_latex[variable]
if 'xsection' in measurement:
properties.y_axis_title = r'$\frac{1}{\sigma} \frac{d\sigma}{d' + \
variables_latex[variable] + '}$'
else:
properties.y_axis_title = r'$t\bar{t}$ normalisation'
histograms = {'Combine before unfolding': h_combineBefore, 'Combine after unfolding': h_combineAfter}
if add_before_unfolding:
histograms['before unfolding'] = h_before_unfolding
properties.name += '_ext'
properties.has_ratio = False
plot = Plot(histograms, properties)
plot.draw_method = 'errorbar'
compare_histograms(plot)
def compare_unfolding_methods(measurement='normalised_xsection',
add_before_unfolding=False, channel='combined'):
file_template = '/hdfs/TopQuarkGroup/run2/dpsData/'
file_template += 'data/normalisation/background_subtraction/13TeV/'
file_template += '{variable}/VisiblePS/central/'
file_template += '{measurement}_{channel}_RooUnfold{method}.txt'
variables = ['MET', 'HT', 'ST', 'NJets',
'lepton_pt', 'abs_lepton_eta', 'WPT']
for variable in variables:
svd = file_template.format(
variable=variable,
method='Svd',
channel=channel,
measurement=measurement)
bayes = file_template.format(
variable=variable,
method='Bayes', channel=channel,
measurement=measurement)
data = read_data_from_JSON(svd)
before_unfolding = data['TTJet_measured_withoutFakes']
svd_data = data['TTJet_unfolded']
bayes_data = read_data_from_JSON(bayes)['TTJet_unfolded']
h_svd = value_error_tuplelist_to_hist(
svd_data, bin_edges_vis[variable])
h_bayes = value_error_tuplelist_to_hist(
bayes_data, bin_edges_vis[variable])
h_before_unfolding = value_error_tuplelist_to_hist(
before_unfolding, bin_edges_vis[variable])
properties = Histogram_properties()
properties.name = '{0}_compare_unfolding_methods_{1}_{2}'.format(
measurement, variable, channel)
properties.title = 'Comparison of unfolding methods'
properties.path = 'plots'
properties.has_ratio = True
properties.xerr = True
properties.x_limits = (
bin_edges_vis[variable][0], bin_edges_vis[variable][-1])
properties.x_axis_title = variables_latex[variable]
if 'xsection' in measurement:
properties.y_axis_title = r'$\frac{1}{\sigma} \frac{d\sigma}{d' + \
variables_latex[variable] + '}$'
else:
properties.y_axis_title = r'$t\bar{t}$ normalisation'
histograms = {'SVD': h_svd, 'Bayes': h_bayes}
if add_before_unfolding:
histograms['before unfolding'] = h_before_unfolding
properties.name += '_ext'
properties.has_ratio = False
plot = Plot(histograms, properties)
plot.draw_method = 'errorbar'
compare_histograms(plot)
def plot_fit_results(fit_results, initial_values, channel):
global variable, output_folder
title = electron_histogram_title if channel == 'electron' else muon_histogram_title
histogram_properties = Histogram_properties()
histogram_properties.title = title
histogram_properties.x_axis_title = variable + ' [GeV]'
histogram_properties.mc_error = 0.0
histogram_properties.legend_location = 'upper right'
# we will need 4 histograms: TTJet, SingleTop, QCD, V+Jets
for sample in ['TTJet', 'SingleTop', 'QCD', 'V+Jets']:
histograms = {}
# absolute eta measurement as baseline
h_absolute_eta = None
h_before = None
histogram_properties.y_axis_title = 'Fitted number of events for ' + samples_latex[
sample]
for fit_var_input in fit_results.keys():
latex_string = create_latex_string(fit_var_input)
fit_data = fit_results[fit_var_input][sample]
h = value_error_tuplelist_to_hist(fit_data, bin_edges[variable])
if fit_var_input == 'absolute_eta':
h_absolute_eta = h
elif fit_var_input == 'before':
h_before = h
else:
histograms[latex_string] = h
graphs = spread_x(histograms.values(), bin_edges[variable])
for key, graph in zip(histograms.keys(), graphs):
histograms[key] = graph
filename = sample.replace('+', '_') + '_fit_var_comparison_' + channel
histogram_properties.name = filename
histogram_properties.y_limits = 0, limit_range_y(
h_absolute_eta)[1] * 1.3
histogram_properties.x_limits = bin_edges[variable][0], bin_edges[
variable][-1]
h_initial_values = value_error_tuplelist_to_hist(
initial_values[sample], bin_edges[variable])
h_initial_values.Scale(closure_tests['simple'][sample])
compare_measurements(models={
fit_variables_latex['absolute_eta']: h_absolute_eta,
'initial values': h_initial_values,
'before': h_before
},
measurements=histograms,
show_measurement_errors=True,
histogram_properties=histogram_properties,
save_folder=output_folder,
save_as=['png', 'pdf'])
def plot_fit_results(fit_results, initial_values, channel):
global variable, output_folder
title = electron_histogram_title if channel == "electron" else muon_histogram_title
histogram_properties = Histogram_properties()
histogram_properties.title = title
histogram_properties.x_axis_title = variable + " [GeV]"
histogram_properties.mc_error = 0.0
histogram_properties.legend_location = "upper right"
# we will need 4 histograms: TTJet, SingleTop, QCD, V+Jets
for sample in ["TTJet", "SingleTop", "QCD", "V+Jets"]:
histograms = {}
# absolute eta measurement as baseline
h_absolute_eta = None
h_before = None
histogram_properties.y_axis_title = "Fitted number of events for " + samples_latex[sample]
for fit_var_input in fit_results.keys():
latex_string = create_latex_string(fit_var_input)
fit_data = fit_results[fit_var_input][sample]
h = value_error_tuplelist_to_hist(fit_data, bin_edges[variable])
if fit_var_input == "absolute_eta":
h_absolute_eta = h
elif fit_var_input == "before":
h_before = h
else:
histograms[latex_string] = h
graphs = spread_x(histograms.values(), bin_edges[variable])
for key, graph in zip(histograms.keys(), graphs):
histograms[key] = graph
filename = sample.replace("+", "_") + "_fit_var_comparison_" + channel
histogram_properties.name = filename
histogram_properties.y_limits = 0, limit_range_y(h_absolute_eta)[1] * 1.3
histogram_properties.x_limits = bin_edges[variable][0], bin_edges[variable][-1]
h_initial_values = value_error_tuplelist_to_hist(initial_values[sample], bin_edges[variable])
h_initial_values.Scale(closure_tests["simple"][sample])
compare_measurements(
models={
fit_variables_latex["absolute_eta"]: h_absolute_eta,
"initial values": h_initial_values,
"before": h_before,
},
measurements=histograms,
show_measurement_errors=True,
histogram_properties=histogram_properties,
save_folder=output_folder,
save_as=["png", "pdf"],
)
def plot_fit_results( fit_results, initial_values, channel ):
global variable, output_folder
title = electron_histogram_title if channel == 'electron' else muon_histogram_title
histogram_properties = Histogram_properties()
histogram_properties.title = title
histogram_properties.x_axis_title = variable + ' [GeV]'
histogram_properties.mc_error = 0.0
histogram_properties.legend_location = 'upper right'
# we will need 4 histograms: TTJet, SingleTop, QCD, V+Jets
for sample in ['TTJet', 'SingleTop', 'QCD', 'V+Jets']:
histograms = {}
# absolute eta measurement as baseline
h_absolute_eta = None
h_before = None
histogram_properties.y_axis_title = 'Fitted number of events for ' + samples_latex[sample]
for fit_var_input in fit_results.keys():
latex_string = create_latex_string( fit_var_input )
fit_data = fit_results[fit_var_input][sample]
h = value_error_tuplelist_to_hist( fit_data,
bin_edges[variable] )
if fit_var_input == 'absolute_eta':
h_absolute_eta = h
elif fit_var_input == 'before':
h_before = h
else:
histograms[latex_string] = h
graphs = spread_x( histograms.values(), bin_edges[variable] )
for key, graph in zip( histograms.keys(), graphs ):
histograms[key] = graph
filename = sample.replace( '+', '_' ) + '_fit_var_comparison_' + channel
histogram_properties.name = filename
histogram_properties.y_limits = 0, limit_range_y( h_absolute_eta )[1] * 1.3
histogram_properties.x_limits = bin_edges[variable][0], bin_edges[variable][-1]
h_initial_values = value_error_tuplelist_to_hist( initial_values[sample],
bin_edges[variable] )
h_initial_values.Scale(closure_tests['simple'][sample])
compare_measurements( models = {fit_variables_latex['absolute_eta']:h_absolute_eta,
'initial values' : h_initial_values,
'before': h_before},
measurements = histograms,
show_measurement_errors = True,
histogram_properties = histogram_properties,
save_folder = output_folder,
save_as = ['png', 'pdf'] )
def read_xsection_measurement_results(category, channel):
global path_to_JSON, variable, k_value, met_type
normalised_xsection_unfolded = None
if category in met_uncertainties and variable == 'HT':
normalised_xsection_unfolded = read_data_from_JSON(path_to_JSON + '/xsection_measurement_results' + '/kv' + str(k_value) + '/'
+ 'central' + '/normalised_xsection_' + channel + '_' + met_type + '.txt')
else:
normalised_xsection_unfolded = read_data_from_JSON(path_to_JSON + '/xsection_measurement_results' + '/kv' + str(k_value) + '/'
+ category + '/normalised_xsection_' + channel + '_' + met_type + '.txt')
h_normalised_xsection = value_error_tuplelist_to_hist(normalised_xsection_unfolded['TTJet_measured'], bin_edges[variable])
h_normalised_xsection_unfolded = value_error_tuplelist_to_hist(normalised_xsection_unfolded['TTJet_unfolded'], bin_edges[variable])
histograms_normalised_xsection_different_generators = {'measured':h_normalised_xsection,
'unfolded':h_normalised_xsection_unfolded}
histograms_normalised_xsection_systematics_shifts = {'measured':h_normalised_xsection,
'unfolded':h_normalised_xsection_unfolded}
if category == 'central':
# true distributions
h_normalised_xsection_MADGRAPH = value_error_tuplelist_to_hist(normalised_xsection_unfolded['MADGRAPH'], bin_edges[variable])
h_normalised_xsection_POWHEG = value_error_tuplelist_to_hist(normalised_xsection_unfolded['POWHEG'], bin_edges[variable])
h_normalised_xsection_MCATNLO = value_error_tuplelist_to_hist(normalised_xsection_unfolded['MCATNLO'], bin_edges[variable])
h_normalised_xsection_mathchingup = value_error_tuplelist_to_hist(normalised_xsection_unfolded['matchingup'], bin_edges[variable])
h_normalised_xsection_mathchingdown = value_error_tuplelist_to_hist(normalised_xsection_unfolded['matchingdown'], bin_edges[variable])
h_normalised_xsection_scaleup = value_error_tuplelist_to_hist(normalised_xsection_unfolded['scaleup'], bin_edges[variable])
h_normalised_xsection_scaledown = value_error_tuplelist_to_hist(normalised_xsection_unfolded['scaledown'], bin_edges[variable])
histograms_normalised_xsection_different_generators.update({'MADGRAPH':h_normalised_xsection_MADGRAPH,
'POWHEG':h_normalised_xsection_POWHEG,
'MCATNLO':h_normalised_xsection_MCATNLO})
histograms_normalised_xsection_systematics_shifts.update({'MADGRAPH':h_normalised_xsection_MADGRAPH,
'matchingdown': h_normalised_xsection_mathchingdown,
'matchingup': h_normalised_xsection_mathchingup,
'scaledown': h_normalised_xsection_scaledown,
'scaleup': h_normalised_xsection_scaleup})
normalised_xsection_unfolded_with_errors = read_data_from_JSON(path_to_JSON + '/xsection_measurement_results' + '/kv' +
str(k_value) + '/' + category + '/normalised_xsection_' +
channel + '_' + met_type + '_with_errors.txt')
# a rootpy.Graph with asymmetric errors!
h_normalised_xsection_with_systematics = value_errors_tuplelist_to_graph(normalised_xsection_unfolded_with_errors['TTJet_measured'], bin_edges[variable])
h_normalised_xsection_with_systematics_unfolded = value_errors_tuplelist_to_graph(normalised_xsection_unfolded_with_errors['TTJet_unfolded'], bin_edges[variable])
histograms_normalised_xsection_different_generators['measured_with_systematics'] = h_normalised_xsection_with_systematics
histograms_normalised_xsection_different_generators['unfolded_with_systematics'] = h_normalised_xsection_with_systematics_unfolded
histograms_normalised_xsection_systematics_shifts['measured_with_systematics'] = h_normalised_xsection_with_systematics
histograms_normalised_xsection_systematics_shifts['unfolded_with_systematics'] = h_normalised_xsection_with_systematics_unfolded
return histograms_normalised_xsection_different_generators, histograms_normalised_xsection_systematics_shifts
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 = bin_edges[variable]
if visiblePS:
edges = bin_edges_vis[variable]
h_data = value_error_tuplelist_to_hist(results, edges)
# Remove fakes before unfolding
h_measured, h_data = removeFakes(h_measured, h_data, h_response)
unfolding = Unfolding(h_truth,
h_measured,
h_response,
h_fakes,
method=method,
k_value=-1,
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(h_data)
del unfolding
return hist_to_value_error_tuplelist(
h_unfolded_data), hist_to_value_error_tuplelist(h_data)
def make_histogram(result, bin_edges):
if len(result[0]) == 2:
h = value_error_tuplelist_to_hist(result, bin_edges)
return h
else: # len(result[0]) == 3
g = value_errors_tuplelist_to_graph(result, bin_edges)
return g
def make_histogram(result, bin_edges):
if len(result[0]) == 2:
h = value_error_tuplelist_to_hist(result, bin_edges)
return h
else: # len(result[0]) == 3
g = value_errors_tuplelist_to_graph(result, bin_edges)
return g
def __set_unfolding_histograms__( self ):
# at the moment only one file is supported for the unfolding input
files = set( [self.truth['file'],
self.gen_vs_reco['file'],
self.measured['file']]
)
if len( files ) > 1:
print "Currently not supported to have different files for truth, gen_vs_reco and measured"
sys.exit()
input_file = files.pop()
visiblePS = False
if self.phaseSpace == 'VisiblePS':
visiblePS = True
t, m, r, f = get_unfold_histogram_tuple( File(input_file),
self.variable,
self.channel,
centre_of_mass = self.centre_of_mass_energy,
ttbar_xsection=self.measurement_config.ttbar_xsection,
luminosity=self.measurement_config.luminosity,
load_fakes = True,
visiblePS = visiblePS
)
self.h_truth = asrootpy ( t )
self.h_response = asrootpy ( r )
self.h_measured = asrootpy ( m )
self.h_fakes = asrootpy ( f )
data_file = self.data['file']
if data_file.endswith('.root'):
self.h_data = get_histogram_from_file(self.data['histogram'], self.data['file'])
elif data_file.endswith('.json') or data_file.endswith('.txt'):
data_key = self.data['histogram']
# assume configured bin edges
edges = []
edges = reco_bin_edges_vis[self.variable]
json_input = read_data_from_JSON(data_file)
if data_key == "": # JSON file == histogram
self.h_data = value_error_tuplelist_to_hist(json_input, edges)
else:
self.h_data = value_error_tuplelist_to_hist(json_input[data_key], edges)
else:
print 'Unkown file extension', data_file.split('.')[-1]
def unfold_results(results, h_truth, h_measured, h_response, method):
global bin_edges
h_data = value_error_tuplelist_to_hist(results, bin_edges)
unfolding = Unfolding(h_truth, h_measured, h_response, method=method)
h_unfolded_data = unfolding.unfold(h_data)
return hist_to_value_error_tuplelist(h_unfolded_data)
def get_data_histogram( channel, variable, met_type ):
fit_result_input = 'data/M3_angle_bl/13TeV/%(variable)s/fit_results/central/fit_results_%(channel)s_%(met_type)s.txt'
fit_results = read_data_from_JSON( fit_result_input % {'channel': channel, 'variable': variable, 'met_type':met_type} )
fit_data = fit_results['TTJet']
print fit_data
print bin_edges[variable]
h_data = value_error_tuplelist_to_hist( fit_data, bin_edges[variable] )
return h_data
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 json_to_histograms(results, channel, variable, category):
global bin_edges
histograms = {}
for measurement, result in results.iteritems():
histograms[measurement] = value_error_tuplelist_to_hist(result, bin_edges[variable])
name = get_name(channel, variable, measurement, category)
histograms[measurement].SetName(name)
return histograms
def plot_pull_from_list(hist_data, hist_min_x,hist_max_x, hist_n_bins):
stats = 19596500
bin_width = (2.0 * hist_max_x) / hist_n_bins
print hist_n_bins, bin_width
bin_edges = list(drange(hist_min_x, hist_max_x, bin_width))
print bin_edges
print len(bin_edges)
h_pull = value_error_tuplelist_to_hist(hist_data, bin_edges)
plot_h_pull(h_pull, stats = stats, name = 'pull_from_list' )
def plot_pull_from_list(hist_data, hist_min_x, hist_max_x, hist_n_bins):
stats = 19596500
bin_width = (2.0 * hist_max_x) / hist_n_bins
print(hist_n_bins, bin_width)
bin_edges = list(drange(hist_min_x, hist_max_x, bin_width))
print(bin_edges)
print(len(bin_edges))
h_pull = value_error_tuplelist_to_hist(hist_data, bin_edges)
plot_h_pull(h_pull, stats=stats, name='pull_from_list')
def json_to_histograms(results, channel, variable, category):
global bin_edges
histograms = {}
for measurement, result in results.iteritems():
histograms[measurement] = value_error_tuplelist_to_hist(
result, bin_edges[variable])
name = get_name(channel, variable, measurement, category)
histograms[measurement].SetName(name)
return histograms
def __set_unfolding_histograms__(self):
# at the moment only one file is supported for the unfolding input
files = set([
self.truth['file'], self.gen_vs_reco['file'], self.measured['file']
])
if len(files) > 1:
print "Currently not supported to have different files for truth, gen_vs_reco and measured"
sys.exit()
input_file = files.pop()
t, m, r, _ = get_unfold_histogram_tuple(
File(input_file),
self.variable,
self.channel,
centre_of_mass=self.centre_of_mass_energy,
ttbar_xsection=self.measurement_config.ttbar_xsection,
luminosity=self.measurement_config.luminosity,
)
self.h_truth = t
self.h_response = r
self.h_measured = m
data_file = self.data['file']
if data_file.endswith('.root'):
self.h_data = get_histogram_from_file(self.data['histogram'],
self.data['file'])
elif data_file.endswith('.json') or data_file.endswith('.txt'):
data_key = self.data['histogram']
# assume configured bin edges
edges = []
if self.phaseSpace == 'FullPS':
edges = bin_edges[self.variable]
elif self.phaseSpace == 'VisiblePS':
edges = bin_edges_vis[self.variable]
json_input = read_data_from_JSON(data_file)
if data_key == "": # JSON file == histogram
self.h_data = value_error_tuplelist_to_hist(json_input, edges)
else:
self.h_data = value_error_tuplelist_to_hist(
json_input[data_key], edges)
else:
print 'Unkown file extension', data_file.split('.')[-1]
def unfold_results( results, category, channel, k_value, h_truth, h_measured, h_response, h_fakes, method ):
global variable, path_to_JSON, options
h_data = value_error_tuplelist_to_hist( results, bin_edges[variable] )
unfolding = Unfolding( h_truth, h_measured, h_response, h_fakes, method = method, k_value = k_value )
# turning off the unfolding errors for systematic samples
if not category == 'central':
unfoldCfg.Hreco = 0
else:
unfoldCfg.Hreco = options.Hreco
h_unfolded_data = unfolding.unfold( h_data )
if options.write_unfolding_objects:
# export the D and SV distributions
SVD_path = path_to_JSON + '/unfolding_objects/' + channel + '/kv_' + str( k_value ) + '/'
make_folder_if_not_exists( SVD_path )
if method == 'TSVDUnfold':
SVDdist = File( 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 = File( 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 = File( 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 read_xsection_measurement_results(category, channel):
global path_to_JSON, variable, k_value, met_type
normalised_xsection_unfolded = read_data_from_JSON(path_to_JSON + '/xsection_measurement_results' + '/kv' + str(k_value) + '/'
+ category + '/normalised_xsection_' + channel + '_' + met_type + '.txt')
h_normalised_xsection = value_error_tuplelist_to_hist(normalised_xsection_unfolded['TTJet_measured'], bin_edges[variable])
h_normalised_xsection_unfolded = value_error_tuplelist_to_hist(normalised_xsection_unfolded['TTJet_unfolded'], bin_edges[variable])
if category == 'central':
normalised_xsection_unfolded_with_errors = read_data_from_JSON(path_to_JSON + '/xsection_measurement_results' + '/kv' +
str(k_value) + '/' + category + '/normalised_xsection_' +
channel + '_' + met_type + '_with_errors.txt')
h_normalised_xsection = value_errors_tuplelist_to_graph(normalised_xsection_unfolded_with_errors['TTJet_measured'], bin_edges[variable])
h_normalised_xsection_unfolded = value_errors_tuplelist_to_graph(normalised_xsection_unfolded_with_errors['TTJet_unfolded'], bin_edges[variable])
#true distributions
h_normalised_xsection_MADGRAPH = value_error_tuplelist_to_hist(normalised_xsection_unfolded['MADGRAPH'], bin_edges[variable])
h_normalised_xsection_POWHEG = value_error_tuplelist_to_hist(normalised_xsection_unfolded['POWHEG'], bin_edges[variable])
h_normalised_xsection_MCATNLO = value_error_tuplelist_to_hist(normalised_xsection_unfolded['MCATNLO'], bin_edges[variable])
h_normalised_xsection_mathchingup = value_error_tuplelist_to_hist(normalised_xsection_unfolded['matchingup'], bin_edges[variable])
h_normalised_xsection_mathchingdown = value_error_tuplelist_to_hist(normalised_xsection_unfolded['matchingdown'], bin_edges[variable])
h_normalised_xsection_scaleup = value_error_tuplelist_to_hist(normalised_xsection_unfolded['scaleup'], bin_edges[variable])
h_normalised_xsection_scaledown = value_error_tuplelist_to_hist(normalised_xsection_unfolded['scaledown'], bin_edges[variable])
histograms_normalised_xsection_different_generators = {
'measured':h_normalised_xsection,
'unfolded':h_normalised_xsection_unfolded,
'MADGRAPH':h_normalised_xsection_MADGRAPH,
'POWHEG':h_normalised_xsection_POWHEG,
'MCATNLO':h_normalised_xsection_MCATNLO
}
histograms_normalised_xsection_systematics_shifts = {
'measured':h_normalised_xsection,
'unfolded':h_normalised_xsection_unfolded,
'matchingdown': h_normalised_xsection_mathchingdown,
'matchingup': h_normalised_xsection_mathchingup,
'scaledown': h_normalised_xsection_scaledown,
'scaleup': h_normalised_xsection_scaleup
}
return histograms_normalised_xsection_different_generators, histograms_normalised_xsection_systematics_shifts
def read_xsection_measurement_results(category, channel):
global path_to_JSON, variable, k_value, met_type
normalised_xsection_unfolded = read_data_from_JSON(
path_to_JSON + '/' + variable + '/xsection_measurement_results' +
'/kv' + str(k_value) + '/' + category + '/normalised_xsection_' +
channel + '_' + met_type + '.txt')
h_normalised_xsection = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['TTJet_measured'], bin_edges[variable])
h_normalised_xsection_unfolded = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['TTJet_unfolded'], bin_edges[variable])
h_normalised_xsection_MADGRAPH = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['MADGRAPH'], bin_edges[variable])
h_normalised_xsection_POWHEG = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['POWHEG'], bin_edges[variable])
h_normalised_xsection_MCATNLO = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['MCATNLO'], bin_edges[variable])
h_normalised_xsection_mathchingup = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['matchingup'], bin_edges[variable])
h_normalised_xsection_mathchingdown = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['matchingdown'], bin_edges[variable])
h_normalised_xsection_scaleup = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['scaleup'], bin_edges[variable])
h_normalised_xsection_scaledown = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['scaledown'], bin_edges[variable])
histograms_normalised_xsection_different_generators = {
'measured': h_normalised_xsection,
'unfolded': h_normalised_xsection_unfolded,
'MADGRAPH': h_normalised_xsection_MADGRAPH,
'POWHEG': h_normalised_xsection_POWHEG,
'MCATNLO': h_normalised_xsection_MCATNLO
}
histograms_normalised_xsection_systematics_shifts = {
'measured': h_normalised_xsection,
'unfolded': h_normalised_xsection_unfolded,
'matchingdown': h_normalised_xsection_mathchingdown,
'matchingup': h_normalised_xsection_mathchingup,
'scaledown': h_normalised_xsection_scaledown,
'scaleup': h_normalised_xsection_scaleup
}
return histograms_normalised_xsection_different_generators, histograms_normalised_xsection_systematics_shifts
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 = bin_edges[variable]
if visiblePS:
edges = bin_edges_vis[variable]
h_data = value_error_tuplelist_to_hist( results, edges )
# Remove fakes before unfolding
h_measured, h_data = removeFakes( h_measured, h_data, h_response )
unfolding = Unfolding( h_truth, h_measured, h_response, h_fakes, method = method, k_value = -1, 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( h_data )
del unfolding
return hist_to_value_error_tuplelist( h_unfolded_data ), hist_to_value_error_tuplelist( h_data )
def debug_last_bin():
'''
For debugging why the last bin in the problematic variables deviates a
lot in _one_ of the channels only.
'''
file_template = '/hdfs/TopQuarkGroup/run2/dpsData/'
file_template += 'data/normalisation/background_subtraction/13TeV/'
file_template += '{variable}/VisiblePS/central/'
file_template += 'normalised_xsection_{channel}_RooUnfoldSvd{suffix}.txt'
problematic_variables = ['HT', 'MET', 'NJets', 'lepton_pt']
for variable in problematic_variables:
results = {}
Result = namedtuple(
'Result', ['before_unfolding', 'after_unfolding', 'model'])
for channel in ['electron', 'muon', 'combined']:
input_file_data = file_template.format(
variable=variable,
channel=channel,
suffix='_with_errors',
)
input_file_model = file_template.format(
variable=variable,
channel=channel,
suffix='',
)
data = read_data_from_JSON(input_file_data)
data_model = read_data_from_JSON(input_file_model)
before_unfolding = data['TTJet_measured_withoutFakes']
after_unfolding = data['TTJet_unfolded']
model = data_model['powhegPythia8']
# only use the last bin
h_before_unfolding = value_errors_tuplelist_to_graph(
[before_unfolding[-1]], bin_edges_vis[variable][-2:])
h_after_unfolding = value_errors_tuplelist_to_graph(
[after_unfolding[-1]], bin_edges_vis[variable][-2:])
h_model = value_error_tuplelist_to_hist(
[model[-1]], bin_edges_vis[variable][-2:])
r = Result(before_unfolding, after_unfolding, model)
h = Result(h_before_unfolding, h_after_unfolding, h_model)
results[channel] = (r, h)
models = {'POWHEG+PYTHIA': results['combined'][1].model}
h_unfolded = [results[channel][1].after_unfolding for channel in [
'electron', 'muon', 'combined']]
tmp_hists = spread_x(h_unfolded, bin_edges_vis[variable][-2:])
measurements = {}
for channel, hist in zip(['electron', 'muon', 'combined'], tmp_hists):
value = results[channel][0].after_unfolding[-1][0]
error = results[channel][0].after_unfolding[-1][1]
label = '{c_label} ({value:1.2g} $\pm$ {error:1.2g})'.format(
c_label=channel,
value=value,
error=error,
)
measurements[label] = hist
properties = Histogram_properties()
properties.name = 'normalised_xsection_compare_channels_{0}_{1}_last_bin'.format(
variable, channel)
properties.title = 'Comparison of channels'
properties.path = 'plots'
properties.has_ratio = True
properties.xerr = False
properties.x_limits = (
bin_edges_vis[variable][-2], bin_edges_vis[variable][-1])
properties.x_axis_title = variables_latex[variable]
properties.y_axis_title = r'$\frac{1}{\sigma} \frac{d\sigma}{d' + \
variables_latex[variable] + '}$'
properties.legend_location = (0.95, 0.40)
if variable == 'NJets':
properties.legend_location = (0.97, 0.80)
properties.formats = ['png']
compare_measurements(models=models, measurements=measurements, show_measurement_errors=True,
histogram_properties=properties, save_folder='plots/', save_as=properties.formats)
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])
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')
def do_shape_check(channel, control_region_1, control_region_2, variable, normalisation, title, x_title, y_title, x_limits, y_limits,
name_region_1='conversions' , name_region_2='non-isolated electrons', name_region_3='fit results', rebin=1):
global b_tag_bin
# QCD shape comparison
if channel == 'electron':
histograms = get_histograms_from_files([control_region_1, control_region_2], histogram_files)
region_1 = histograms[channel][control_region_1].Clone() - histograms['TTJet'][control_region_1].Clone() - histograms['V+Jets'][control_region_1].Clone() - histograms['SingleTop'][control_region_1].Clone()
region_2 = histograms[channel][control_region_2].Clone() - histograms['TTJet'][control_region_2].Clone() - histograms['V+Jets'][control_region_2].Clone() - histograms['SingleTop'][control_region_2].Clone()
region_1.Rebin(rebin)
region_2.Rebin(rebin)
histogram_properties = Histogram_properties()
histogram_properties.name = 'QCD_control_region_comparison_' + channel + '_' + variable + '_' + b_tag_bin
histogram_properties.title = title + ', ' + b_tag_bins_latex[b_tag_bin]
histogram_properties.x_axis_title = x_title
histogram_properties.y_axis_title = 'arbitrary units/(0.1)'
histogram_properties.x_limits = x_limits
histogram_properties.y_limits = y_limits[0]
histogram_properties.mc_error = 0.0
histogram_properties.legend_location = 'upper right'
make_control_region_comparison(region_1, region_2,
name_region_1=name_region_1, name_region_2=name_region_2,
histogram_properties=histogram_properties, save_folder=output_folder)
# QCD shape comparison to fit results
histograms = get_histograms_from_files([control_region_1], histogram_files)
region_1_tmp = histograms[channel][control_region_1].Clone() - histograms['TTJet'][control_region_1].Clone() - histograms['V+Jets'][control_region_1].Clone() - histograms['SingleTop'][control_region_1].Clone()
region_1 = rebin_asymmetric(region_1_tmp, bin_edges[variable])
fit_results_QCD = normalisation[variable]['QCD']
region_2 = value_error_tuplelist_to_hist(fit_results_QCD, bin_edges_vis[variable])
histogram_properties = Histogram_properties()
histogram_properties.name = 'QCD_control_region_comparison_' + channel + '_' + variable + '_fits_with_conversions_' + b_tag_bin
histogram_properties.title = title + ', ' + b_tag_bins_latex[b_tag_bin]
histogram_properties.x_axis_title = x_title
histogram_properties.y_axis_title = 'arbitrary units/(0.1)'
histogram_properties.x_limits = x_limits
histogram_properties.y_limits = y_limits[1]
histogram_properties.mc_error = 0.0
histogram_properties.legend_location = 'upper right'
make_control_region_comparison(region_1, region_2,
name_region_1=name_region_1, name_region_2=name_region_3,
histogram_properties=histogram_properties, save_folder=output_folder)
histograms = get_histograms_from_files([control_region_2], histogram_files)
region_1_tmp = histograms[channel][control_region_2].Clone() - histograms['TTJet'][control_region_2].Clone() - histograms['V+Jets'][control_region_2].Clone() - histograms['SingleTop'][control_region_2].Clone()
region_1 = rebin_asymmetric(region_1_tmp, bin_edges_vis[variable])
fit_results_QCD = normalisation[variable]['QCD']
region_2 = value_error_tuplelist_to_hist(fit_results_QCD, bin_edges[variable])
histogram_properties = Histogram_properties()
histogram_properties.name = 'QCD_control_region_comparison_' + channel + '_' + variable + '_fits_with_noniso_' + b_tag_bin
histogram_properties.title = title + ', ' + b_tag_bins_latex[b_tag_bin]
histogram_properties.x_axis_title = x_title
histogram_properties.y_axis_title = 'arbitrary units/(0.1)'
histogram_properties.x_limits = x_limits
histogram_properties.y_limits = y_limits[1]
histogram_properties.mc_error = 0.0
histogram_properties.legend_location = 'upper right'
make_control_region_comparison(region_1, region_2,
name_region_1=name_region_2, name_region_2=name_region_3,
histogram_properties=histogram_properties, save_folder=output_folder)
def do_shape_check(
channel,
control_region_1,
control_region_2,
variable,
normalisation,
title,
x_title,
y_title,
x_limits,
y_limits,
name_region_1="conversions",
name_region_2="non-isolated electrons",
name_region_3="fit results",
rebin=1,
):
global b_tag_bin
# QCD shape comparison
if channel == "electron":
histograms = get_histograms_from_files([control_region_1, control_region_2], histogram_files)
region_1 = (
histograms[channel][control_region_1].Clone()
- histograms["TTJet"][control_region_1].Clone()
- histograms["V+Jets"][control_region_1].Clone()
- histograms["SingleTop"][control_region_1].Clone()
)
region_2 = (
histograms[channel][control_region_2].Clone()
- histograms["TTJet"][control_region_2].Clone()
- histograms["V+Jets"][control_region_2].Clone()
- histograms["SingleTop"][control_region_2].Clone()
)
region_1.Rebin(rebin)
region_2.Rebin(rebin)
histogram_properties = Histogram_properties()
histogram_properties.name = "QCD_control_region_comparison_" + channel + "_" + variable + "_" + b_tag_bin
histogram_properties.title = title + ", " + b_tag_bins_latex[b_tag_bin]
histogram_properties.x_axis_title = x_title
histogram_properties.y_axis_title = "arbitrary units/(0.1)"
histogram_properties.x_limits = x_limits
histogram_properties.y_limits = y_limits[0]
histogram_properties.mc_error = 0.0
histogram_properties.legend_location = "upper right"
make_control_region_comparison(
region_1,
region_2,
name_region_1=name_region_1,
name_region_2=name_region_2,
histogram_properties=histogram_properties,
save_folder=output_folder,
)
# QCD shape comparison to fit results
histograms = get_histograms_from_files([control_region_1], histogram_files)
region_1_tmp = (
histograms[channel][control_region_1].Clone()
- histograms["TTJet"][control_region_1].Clone()
- histograms["V+Jets"][control_region_1].Clone()
- histograms["SingleTop"][control_region_1].Clone()
)
region_1 = rebin_asymmetric(region_1_tmp, bin_edges[variable])
fit_results_QCD = normalisation[variable]["QCD"]
region_2 = value_error_tuplelist_to_hist(fit_results_QCD, bin_edges[variable])
histogram_properties = Histogram_properties()
histogram_properties.name = (
"QCD_control_region_comparison_" + channel + "_" + variable + "_fits_with_conversions_" + b_tag_bin
)
histogram_properties.title = title + ", " + b_tag_bins_latex[b_tag_bin]
histogram_properties.x_axis_title = x_title
histogram_properties.y_axis_title = "arbitrary units/(0.1)"
histogram_properties.x_limits = x_limits
histogram_properties.y_limits = y_limits[1]
histogram_properties.mc_error = 0.0
histogram_properties.legend_location = "upper right"
make_control_region_comparison(
region_1,
region_2,
name_region_1=name_region_1,
name_region_2=name_region_3,
histogram_properties=histogram_properties,
save_folder=output_folder,
)
histograms = get_histograms_from_files([control_region_2], histogram_files)
region_1_tmp = (
histograms[channel][control_region_2].Clone()
- histograms["TTJet"][control_region_2].Clone()
- histograms["V+Jets"][control_region_2].Clone()
- histograms["SingleTop"][control_region_2].Clone()
)
region_1 = rebin_asymmetric(region_1_tmp, bin_edges[variable])
fit_results_QCD = normalisation[variable]["QCD"]
region_2 = value_error_tuplelist_to_hist(fit_results_QCD, bin_edges[variable])
histogram_properties = Histogram_properties()
histogram_properties.name = (
"QCD_control_region_comparison_" + channel + "_" + variable + "_fits_with_noniso_" + b_tag_bin
)
histogram_properties.title = title + ", " + b_tag_bins_latex[b_tag_bin]
histogram_properties.x_axis_title = x_title
histogram_properties.y_axis_title = "arbitrary units/(0.1)"
histogram_properties.x_limits = x_limits
histogram_properties.y_limits = y_limits[1]
histogram_properties.mc_error = 0.0
histogram_properties.legend_location = "upper right"
make_control_region_comparison(
region_1,
region_2,
name_region_1=name_region_2,
name_region_2=name_region_3,
histogram_properties=histogram_properties,
save_folder=output_folder,
)
def test_get_max_y_hist():
h = value_error_tuplelist_to_hist(data_h, bin_edges)
max_y = get_best_max_y([h])
assert max_y == 3 + 1
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')
def debug_last_bin():
'''
For debugging why the last bin in the problematic variables deviates a
lot in _one_ of the channels only.
'''
file_template = '/hdfs/TopQuarkGroup/run2/dpsData/'
file_template += 'data/normalisation/background_subtraction/13TeV/'
file_template += '{variable}/VisiblePS/central/'
file_template += 'normalised_xsection_{channel}_RooUnfoldSvd{suffix}.txt'
problematic_variables = ['HT', 'MET', 'NJets', 'lepton_pt']
for variable in problematic_variables:
results = {}
Result = namedtuple('Result',
['before_unfolding', 'after_unfolding', 'model'])
for channel in ['electron', 'muon', 'combined']:
input_file_data = file_template.format(
variable=variable,
channel=channel,
suffix='_with_errors',
)
input_file_model = file_template.format(
variable=variable,
channel=channel,
suffix='',
)
data = read_data_from_JSON(input_file_data)
data_model = read_data_from_JSON(input_file_model)
before_unfolding = data['TTJet_measured_withoutFakes']
after_unfolding = data['TTJet_unfolded']
model = data_model['powhegPythia8']
# only use the last bin
h_before_unfolding = value_errors_tuplelist_to_graph(
[before_unfolding[-1]], bin_edges_vis[variable][-2:])
h_after_unfolding = value_errors_tuplelist_to_graph(
[after_unfolding[-1]], bin_edges_vis[variable][-2:])
h_model = value_error_tuplelist_to_hist(
[model[-1]], bin_edges_vis[variable][-2:])
r = Result(before_unfolding, after_unfolding, model)
h = Result(h_before_unfolding, h_after_unfolding, h_model)
results[channel] = (r, h)
models = {'POWHEG+PYTHIA': results['combined'][1].model}
h_unfolded = [
results[channel][1].after_unfolding
for channel in ['electron', 'muon', 'combined']
]
tmp_hists = spread_x(h_unfolded, bin_edges_vis[variable][-2:])
measurements = {}
for channel, hist in zip(['electron', 'muon', 'combined'], tmp_hists):
value = results[channel][0].after_unfolding[-1][0]
error = results[channel][0].after_unfolding[-1][1]
label = '{c_label} ({value:1.2g} $\pm$ {error:1.2g})'.format(
c_label=channel,
value=value,
error=error,
)
measurements[label] = hist
properties = Histogram_properties()
properties.name = 'normalised_xsection_compare_channels_{0}_{1}_last_bin'.format(
variable, channel)
properties.title = 'Comparison of channels'
properties.path = 'plots'
properties.has_ratio = True
properties.xerr = False
properties.x_limits = (bin_edges_vis[variable][-2],
bin_edges_vis[variable][-1])
properties.x_axis_title = variables_latex[variable]
properties.y_axis_title = r'$\frac{1}{\sigma} \frac{d\sigma}{d' + \
variables_latex[variable] + '}$'
properties.legend_location = (0.95, 0.40)
if variable == 'NJets':
properties.legend_location = (0.97, 0.80)
properties.formats = ['png']
compare_measurements(models=models,
measurements=measurements,
show_measurement_errors=True,
histogram_properties=properties,
save_folder='plots/',
save_as=properties.formats)
def test_invalid_zero_data( self ):
variable = 'MET'
channel = 'electron'
pseudo_data = value_error_tuplelist_to_hist( [(0,0)]*( len( bin_edges[variable] ) - 1 ), bin_edges[variable] )
self.assertRaises(ValueError, self.dict[channel][variable]['unfolding_object'].unfold, (pseudo_data))
def read_xsection_measurement_results( path_to_JSON, variable, bin_edges,
category,
channel,
k_values,
met_type = 'patType1CorrectedPFMet',
met_uncertainties = [] ):
filename = ''
if category in met_uncertainties and variable == 'HT':
filename = path_to_JSON + '/xsection_measurement_results/' + channel + '/kv' + str( k_values[channel] ) + '/central/normalised_xsection_' + met_type + '.txt'
else:
filename = path_to_JSON + '/xsection_measurement_results/' + channel + '/kv' + str( k_values[channel] ) + '/' + category + '/normalised_xsection_' + met_type + '.txt'
if channel == 'combined':
filename = filename.replace( 'kv' + str( k_values[channel] ), '' )
normalised_xsection_unfolded = read_data_from_JSON( filename )
h_normalised_xsection = value_error_tuplelist_to_hist( normalised_xsection_unfolded['TTJet_measured'], bin_edges[variable] )
h_normalised_xsection_unfolded = value_error_tuplelist_to_hist( normalised_xsection_unfolded['TTJet_unfolded'], bin_edges[variable] )
histograms_normalised_xsection_different_generators = {'measured':h_normalised_xsection,
'unfolded':h_normalised_xsection_unfolded}
histograms_normalised_xsection_systematics_shifts = {'measured':h_normalised_xsection,
'unfolded':h_normalised_xsection_unfolded}
if category == 'central':
# true distributions
h_normalised_xsection_MADGRAPH = value_error_tuplelist_to_hist( normalised_xsection_unfolded['MADGRAPH'], bin_edges[variable] )
h_normalised_xsection_POWHEG_PYTHIA = value_error_tuplelist_to_hist( normalised_xsection_unfolded['POWHEG_PYTHIA'], bin_edges[variable] )
h_normalised_xsection_POWHEG_HERWIG = value_error_tuplelist_to_hist( normalised_xsection_unfolded['POWHEG_HERWIG'], bin_edges[variable] )
h_normalised_xsection_MCATNLO = value_error_tuplelist_to_hist( normalised_xsection_unfolded['MCATNLO'], bin_edges[variable] )
h_normalised_xsection_mathchingup = value_error_tuplelist_to_hist( normalised_xsection_unfolded['matchingup'], bin_edges[variable] )
h_normalised_xsection_mathchingdown = value_error_tuplelist_to_hist( normalised_xsection_unfolded['matchingdown'], bin_edges[variable] )
h_normalised_xsection_scaleup = value_error_tuplelist_to_hist( normalised_xsection_unfolded['scaleup'], bin_edges[variable] )
h_normalised_xsection_scaledown = value_error_tuplelist_to_hist( normalised_xsection_unfolded['scaledown'], bin_edges[variable] )
histograms_normalised_xsection_different_generators.update( {'MADGRAPH':h_normalised_xsection_MADGRAPH,
'POWHEG_PYTHIA':h_normalised_xsection_POWHEG_PYTHIA,
'POWHEG_HERWIG':h_normalised_xsection_POWHEG_HERWIG,
'MCATNLO':h_normalised_xsection_MCATNLO} )
histograms_normalised_xsection_systematics_shifts.update( {'MADGRAPH':h_normalised_xsection_MADGRAPH,
'matchingdown': h_normalised_xsection_mathchingdown,
'matchingup': h_normalised_xsection_mathchingup,
'scaledown': h_normalised_xsection_scaledown,
'scaleup': h_normalised_xsection_scaleup} )
file_template = path_to_JSON + '/xsection_measurement_results/' + channel + '/kv' + str( k_values[channel] ) + '/' + category + '/normalised_xsection_' + met_type
if channel == 'combined':
file_template = file_template.replace( 'kv' + str( k_values[channel] ), '' )
# normalised_xsection_unfolded_with_errors = read_data_from_JSON( file_template + '_with_errors.txt' )
normalised_xsection_unfolded_with_errors_with_systematics_but_without_ttbar_theory = read_data_from_JSON( file_template + '_with_systematics_but_without_ttbar_theory_errors.txt' )
normalised_xsection_unfolded_with_errors_with_systematics_but_without_generator = read_data_from_JSON( file_template + '_with_systematics_but_without_generator_errors.txt' )
# a rootpy.Graph with asymmetric errors!
h_normalised_xsection_with_systematics_but_without_ttbar_theory = value_errors_tuplelist_to_graph(
normalised_xsection_unfolded_with_errors_with_systematics_but_without_ttbar_theory['TTJet_measured'],
bin_edges[variable] )
h_normalised_xsection_with_systematics_but_without_ttbar_theory_unfolded = value_errors_tuplelist_to_graph(
normalised_xsection_unfolded_with_errors_with_systematics_but_without_ttbar_theory['TTJet_unfolded'],
bin_edges[variable] )
h_normalised_xsection_with_systematics_but_without_generator = value_errors_tuplelist_to_graph(
normalised_xsection_unfolded_with_errors_with_systematics_but_without_generator['TTJet_measured'],
bin_edges[variable] )
h_normalised_xsection_with_systematics_but_without_generator_unfolded = value_errors_tuplelist_to_graph(
normalised_xsection_unfolded_with_errors_with_systematics_but_without_generator['TTJet_unfolded'],
bin_edges[variable] )
histograms_normalised_xsection_different_generators['measured_with_systematics'] = h_normalised_xsection_with_systematics_but_without_generator
histograms_normalised_xsection_different_generators['unfolded_with_systematics'] = h_normalised_xsection_with_systematics_but_without_generator_unfolded
histograms_normalised_xsection_systematics_shifts['measured_with_systematics'] = h_normalised_xsection_with_systematics_but_without_ttbar_theory
histograms_normalised_xsection_systematics_shifts['unfolded_with_systematics'] = h_normalised_xsection_with_systematics_but_without_ttbar_theory_unfolded
return histograms_normalised_xsection_different_generators, histograms_normalised_xsection_systematics_shifts
def do_shape_check(channel,
control_region_1,
control_region_2,
variable,
normalisation,
title,
x_title,
y_title,
x_limits,
y_limits,
name_region_1='conversions',
name_region_2='non-isolated electrons',
name_region_3='fit results',
rebin=1):
global b_tag_bin
# QCD shape comparison
if channel == 'electron':
histograms = get_histograms_from_files(
[control_region_1, control_region_2], histogram_files)
region_1 = histograms[channel][control_region_1].Clone(
) - histograms['TTJet'][control_region_1].Clone(
) - histograms['V+Jets'][control_region_1].Clone(
) - histograms['SingleTop'][control_region_1].Clone()
region_2 = histograms[channel][control_region_2].Clone(
) - histograms['TTJet'][control_region_2].Clone(
) - histograms['V+Jets'][control_region_2].Clone(
) - histograms['SingleTop'][control_region_2].Clone()
region_1.Rebin(rebin)
region_2.Rebin(rebin)
histogram_properties = Histogram_properties()
histogram_properties.name = 'QCD_control_region_comparison_' + channel + '_' + variable + '_' + b_tag_bin
histogram_properties.title = title + ', ' + b_tag_bins_latex[b_tag_bin]
histogram_properties.x_axis_title = x_title
histogram_properties.y_axis_title = 'arbitrary units/(0.1)'
histogram_properties.x_limits = x_limits
histogram_properties.y_limits = y_limits[0]
histogram_properties.mc_error = 0.0
histogram_properties.legend_location = 'upper right'
make_control_region_comparison(
region_1,
region_2,
name_region_1=name_region_1,
name_region_2=name_region_2,
histogram_properties=histogram_properties,
save_folder=output_folder)
# QCD shape comparison to fit results
histograms = get_histograms_from_files([control_region_1],
histogram_files)
region_1_tmp = histograms[channel][control_region_1].Clone(
) - histograms['TTJet'][control_region_1].Clone(
) - histograms['V+Jets'][control_region_1].Clone(
) - histograms['SingleTop'][control_region_1].Clone()
region_1 = rebin_asymmetric(region_1_tmp, bin_edges[variable])
fit_results_QCD = normalisation[variable]['QCD']
region_2 = value_error_tuplelist_to_hist(fit_results_QCD,
bin_edges[variable])
histogram_properties = Histogram_properties()
histogram_properties.name = 'QCD_control_region_comparison_' + channel + '_' + variable + '_fits_with_conversions_' + b_tag_bin
histogram_properties.title = title + ', ' + b_tag_bins_latex[b_tag_bin]
histogram_properties.x_axis_title = x_title
histogram_properties.y_axis_title = 'arbitrary units/(0.1)'
histogram_properties.x_limits = x_limits
histogram_properties.y_limits = y_limits[1]
histogram_properties.mc_error = 0.0
histogram_properties.legend_location = 'upper right'
make_control_region_comparison(
region_1,
region_2,
name_region_1=name_region_1,
name_region_2=name_region_3,
histogram_properties=histogram_properties,
save_folder=output_folder)
histograms = get_histograms_from_files([control_region_2], histogram_files)
region_1_tmp = histograms[channel][control_region_2].Clone(
) - histograms['TTJet'][control_region_2].Clone(
) - histograms['V+Jets'][control_region_2].Clone(
) - histograms['SingleTop'][control_region_2].Clone()
region_1 = rebin_asymmetric(region_1_tmp, bin_edges[variable])
fit_results_QCD = normalisation[variable]['QCD']
region_2 = value_error_tuplelist_to_hist(fit_results_QCD,
bin_edges[variable])
histogram_properties = Histogram_properties()
histogram_properties.name = 'QCD_control_region_comparison_' + channel + '_' + variable + '_fits_with_noniso_' + b_tag_bin
histogram_properties.title = title + ', ' + b_tag_bins_latex[b_tag_bin]
histogram_properties.x_axis_title = x_title
histogram_properties.y_axis_title = 'arbitrary units/(0.1)'
histogram_properties.x_limits = x_limits
histogram_properties.y_limits = y_limits[1]
histogram_properties.mc_error = 0.0
histogram_properties.legend_location = 'upper right'
make_control_region_comparison(region_1,
region_2,
name_region_1=name_region_2,
name_region_2=name_region_3,
histogram_properties=histogram_properties,
save_folder=output_folder)
def test_invalid_zero_data( self ):
variable = 'MET'
channel = 'electron'
pseudo_data = value_error_tuplelist_to_hist( [(0,0)]*( len( bin_edges[variable] ) - 1 ), bin_edges[variable] )
self.assertRaises(ValueError, self.dict[channel][variable]['unfolding_object'].unfold, (pseudo_data))
def read_xsection_measurement_results( category, channel ):
global path_to_JSON, variable, met_type, phase_space, method
file_template = '{path}/{category}/{name}_{channel}_{method}{suffix}.txt'
filename = file_template.format(
path = path_to_JSON,
category = category,
name = 'normalised_xsection',
channel = channel,
method = method,
suffix = '',
)
xsec_04_log.debug('Reading file {0}'.format(filename))
normalised_xsection_unfolded = read_data_from_JSON( filename )
edges = bin_edges[variable]
if phase_space == 'VisiblePS':
edges = bin_edges_vis[variable]
h_normalised_xsection = value_error_tuplelist_to_hist( normalised_xsection_unfolded['TTJet_measured'], edges )
h_normalised_xsection_unfolded = value_error_tuplelist_to_hist( normalised_xsection_unfolded['TTJet_unfolded'], edges )
histograms_normalised_xsection_different_generators = {'measured':h_normalised_xsection,
'unfolded':h_normalised_xsection_unfolded}
histograms_normalised_xsection_systematics_shifts = {'measured':h_normalised_xsection,
'unfolded':h_normalised_xsection_unfolded}
if category == 'central':
# true distributions
h_normalised_xsection_powhegPythia8 = value_error_tuplelist_to_hist( normalised_xsection_unfolded['powhegPythia8'], edges )
h_normalised_xsection_amcatnlo = value_error_tuplelist_to_hist( normalised_xsection_unfolded['amcatnlo'], edges )
h_normalised_xsection_madgraphMLM = value_error_tuplelist_to_hist( normalised_xsection_unfolded['madgraphMLM'], edges )
h_normalised_xsection_amcatnloHerwigpp = value_error_tuplelist_to_hist( normalised_xsection_unfolded['amcatnlo_HERWIG'], edges )
h_normalised_xsection_scaleup = value_error_tuplelist_to_hist( normalised_xsection_unfolded['scaleup'], edges )
h_normalised_xsection_scaledown = value_error_tuplelist_to_hist( normalised_xsection_unfolded['scaledown'], edges )
h_normalised_xsection_massup = value_error_tuplelist_to_hist( normalised_xsection_unfolded['massup'], edges )
h_normalised_xsection_massdown = value_error_tuplelist_to_hist( normalised_xsection_unfolded['massdown'], edges )
histograms_normalised_xsection_different_generators.update( {
'powhegPythia8':h_normalised_xsection_powhegPythia8,
'amcatnlo':h_normalised_xsection_amcatnlo,
'madgraphMLM':h_normalised_xsection_madgraphMLM,
'amcatnlo_HERWIG':h_normalised_xsection_amcatnloHerwigpp,
})
histograms_normalised_xsection_systematics_shifts.update( {'powhegPythia8':h_normalised_xsection_powhegPythia8,
'scaledown': h_normalised_xsection_scaledown,
'scaleup': h_normalised_xsection_scaleup,
'massdown': h_normalised_xsection_massdown,
'massup': h_normalised_xsection_massup
})
filename = file_template.format(
path = path_to_JSON,
category = category,
name = 'normalised_xsection',
channel = channel,
method = method,
suffix = '_with_errors',
)
normalised_xsection_unfolded_with_errors = read_data_from_JSON( filename )
xsec_04_log.debug('Reading file {0}'.format(filename))
# filename = file_template.format(
# path = path_to_JSON,
# category = category,
# name = 'normalised_xsection',
# channel = channel,
# method = method,
# suffix = '_with_systematics_but_without_generator_errors',
# )
### normalised_xsection_unfolded_with_errors_with_systematics_but_without_ttbar_theory = read_data_from_JSON( file_template + '_with_systematics_but_without_ttbar_theory_errors.txt' )
# normalised_xsection_unfolded_with_errors_with_systematics_but_without_generator = normalised_xsection_unfolded_with_errors
# a rootpy.Graph with asymmetric errors!
### h_normalised_xsection_with_systematics_but_without_ttbar_theory = value_errors_tuplelist_to_graph(
### normalised_xsection_unfolded_with_errors_with_systematics_but_without_ttbar_theory['TTJet_measured'],
### edges )
### h_normalised_xsection_with_systematics_but_without_ttbar_theory_unfolded = value_errors_tuplelist_to_graph(
### normalised_xsection_unfolded_with_errors_with_systematics_but_without_ttbar_theory['TTJet_unfolded'],
### edges )
h_normalised_xsection_unfolded_with_errors = value_errors_tuplelist_to_graph(
normalised_xsection_unfolded_with_errors['TTJet_measured'],
edges )
h_normalised_xsection_unfolded_with_errors_unfolded = value_errors_tuplelist_to_graph(
normalised_xsection_unfolded_with_errors['TTJet_unfolded'],
edges )
# histograms_normalised_xsection_different_generators['measured_with_systematics'] = h_normalised_xsection_with_systematics_but_without_ttbar_theory
# histograms_normalised_xsection_different_generators['unfolded_with_systematics'] = h_normalised_xsection_with_systematics_but_without_ttbar_theory_unfolded
histograms_normalised_xsection_different_generators['measured_with_systematics'] = h_normalised_xsection_unfolded_with_errors
histograms_normalised_xsection_different_generators['unfolded_with_systematics'] = h_normalised_xsection_unfolded_with_errors_unfolded
histograms_normalised_xsection_systematics_shifts['measured_with_systematics'] = h_normalised_xsection_unfolded_with_errors
histograms_normalised_xsection_systematics_shifts['unfolded_with_systematics'] = h_normalised_xsection_unfolded_with_errors_unfolded
return histograms_normalised_xsection_different_generators, histograms_normalised_xsection_systematics_shifts
def test_get_max_y_hist():
h = value_error_tuplelist_to_hist(data_h, bin_edges)
max_y = get_best_max_y([h])
assert max_y == 3 + 1
def read_xsection_measurement_results(category, channel):
global path_to_JSON, variable, k_value, met_type
normalised_xsection_unfolded = read_data_from_JSON(
path_to_JSON
+ "/"
+ variable
+ "/xsection_measurement_results"
+ "/kv"
+ str(k_value)
+ "/"
+ category
+ "/normalised_xsection_"
+ channel
+ "_"
+ met_type
+ ".txt"
)
h_normalised_xsection = value_error_tuplelist_to_hist(
normalised_xsection_unfolded["TTJet_measured"], bin_edges[variable]
)
h_normalised_xsection_unfolded = value_error_tuplelist_to_hist(
normalised_xsection_unfolded["TTJet_unfolded"], bin_edges[variable]
)
h_normalised_xsection_MADGRAPH = value_error_tuplelist_to_hist(
normalised_xsection_unfolded["MADGRAPH"], bin_edges[variable]
)
h_normalised_xsection_POWHEG = value_error_tuplelist_to_hist(
normalised_xsection_unfolded["POWHEG"], bin_edges[variable]
)
h_normalised_xsection_MCATNLO = value_error_tuplelist_to_hist(
normalised_xsection_unfolded["MCATNLO"], bin_edges[variable]
)
h_normalised_xsection_mathchingup = value_error_tuplelist_to_hist(
normalised_xsection_unfolded["matchingup"], bin_edges[variable]
)
h_normalised_xsection_mathchingdown = value_error_tuplelist_to_hist(
normalised_xsection_unfolded["matchingdown"], bin_edges[variable]
)
h_normalised_xsection_scaleup = value_error_tuplelist_to_hist(
normalised_xsection_unfolded["scaleup"], bin_edges[variable]
)
h_normalised_xsection_scaledown = value_error_tuplelist_to_hist(
normalised_xsection_unfolded["scaledown"], bin_edges[variable]
)
histograms_normalised_xsection_different_generators = {
"measured": h_normalised_xsection,
"unfolded": h_normalised_xsection_unfolded,
"MADGRAPH": h_normalised_xsection_MADGRAPH,
"POWHEG": h_normalised_xsection_POWHEG,
"MCATNLO": h_normalised_xsection_MCATNLO,
}
histograms_normalised_xsection_systematics_shifts = {
"measured": h_normalised_xsection,
"unfolded": h_normalised_xsection_unfolded,
"matchingdown": h_normalised_xsection_mathchingdown,
"matchingup": h_normalised_xsection_mathchingup,
"scaledown": h_normalised_xsection_scaledown,
"scaleup": h_normalised_xsection_scaleup,
}
return histograms_normalised_xsection_different_generators, histograms_normalised_xsection_systematics_shifts
def get_data_histogram( channel, variable, met_type ):
fit_result_input = 'data/M3_angle_bl/13TeV/%(variable)s/fit_results/central/fit_results_%(channel)s_%(met_type)s.txt'
fit_results = read_data_from_JSON( fit_result_input % {'channel': channel, 'variable': variable, 'met_type':met_type} )
fit_data = fit_results['TTJet']
h_data = value_error_tuplelist_to_hist( fit_data, bin_edges[variable] )
return h_data
def main():
config = XSectionConfig(13)
file_for_powhegPythia = File(config.unfolding_central, 'read')
file_for_ptReweight_up = File(config.unfolding_ptreweight_up, 'read')
file_for_ptReweight_down = File(config.unfolding_ptreweight_down, 'read')
file_for_data_template = 'data/normalisation/background_subtraction/13TeV/{variable}/VisiblePS/central/normalisation_combined_patType1CorrectedPFMet.txt'
for channel in ['combined']:
for variable in config.variables:
print variable
# for variable in ['HT']:
# Get the central powheg pythia distributions
_, _, response_central, fakes_central = get_unfold_histogram_tuple(
inputfile=file_for_powhegPythia,
variable=variable,
channel=channel,
centre_of_mass=13,
load_fakes=True,
visiblePS=True
)
measured_central = asrootpy(response_central.ProjectionX('px',1))
truth_central = asrootpy(response_central.ProjectionY())
# Get the reweighted powheg pythia distributions
_, _, response_reweighted_up, _ = get_unfold_histogram_tuple(
inputfile=file_for_ptReweight_up,
variable=variable,
channel=channel,
centre_of_mass=13,
load_fakes=False,
visiblePS=True
)
measured_reweighted_up = asrootpy(response_reweighted_up.ProjectionX('px',1))
truth_reweighted_up = asrootpy(response_reweighted_up.ProjectionY())
_, _, response_reweighted_down, _ = get_unfold_histogram_tuple(
inputfile=file_for_ptReweight_down,
variable=variable,
channel=channel,
centre_of_mass=13,
load_fakes=False,
visiblePS=True
)
measured_reweighted_down = asrootpy(response_reweighted_down.ProjectionX('px',1))
truth_reweighted_down = asrootpy(response_reweighted_down.ProjectionY())
# Get the data input (data after background subtraction, and fake removal)
file_for_data = file_for_data_template.format( variable = variable )
data = read_data_from_JSON(file_for_data)['TTJet']
data = value_error_tuplelist_to_hist( data, reco_bin_edges_vis[variable] )
data = removeFakes( measured_central, fakes_central, data )
# Plot all three
hp = Histogram_properties()
hp.name = 'Reweighting_check_{channel}_{variable}_at_{com}TeV'.format(
channel=channel,
variable=variable,
com='13',
)
v_latex = latex_labels.variables_latex[variable]
unit = ''
if variable in ['HT', 'ST', 'MET', 'WPT', 'lepton_pt']:
unit = ' [GeV]'
hp.x_axis_title = v_latex + unit
hp.y_axis_title = 'Number of events'
hp.title = 'Reweighting check for {variable}'.format(variable=v_latex)
measured_central.Rebin(2)
measured_reweighted_up.Rebin(2)
measured_reweighted_down.Rebin(2)
data.Rebin(2)
measured_central.Scale( 1 / measured_central.Integral() )
measured_reweighted_up.Scale( 1 / measured_reweighted_up.Integral() )
measured_reweighted_down.Scale( 1 / measured_reweighted_down.Integral() )
data.Scale( 1 / data.Integral() )
compare_measurements(
models = {'Central' : measured_central, 'Reweighted Up' : measured_reweighted_up, 'Reweighted Down' : measured_reweighted_down},
measurements = {'Data' : data},
show_measurement_errors=True,
histogram_properties=hp,
save_folder='plots/unfolding/reweighting_check',
save_as=['pdf']
)
def read_xsection_measurement_results(path_to_JSON,
variable,
bin_edges,
category,
channel,
k_values,
met_type='patType1CorrectedPFMet',
met_uncertainties=[]):
filename = ''
if category in met_uncertainties and variable == 'HT' and not 'JES' in category and not 'JER' in category:
filename = path_to_JSON + '/xsection_measurement_results/' + \
channel + '/central/normalised_xsection_' + met_type + '.txt'
else:
filename = path_to_JSON + '/xsection_measurement_results/' + channel + \
'/' + category + '/normalised_xsection_' + met_type + '.txt'
if channel == 'combined':
filename = filename.replace('kv' + str(k_values[channel]), '')
normalised_xsection_unfolded = read_data_from_JSON(filename)
h_normalised_xsection = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['TTJet_measured'], bin_edges[variable])
h_normalised_xsection_unfolded = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['TTJet_unfolded'], bin_edges[variable])
histograms_normalised_xsection_different_generators = {
'measured': h_normalised_xsection,
'unfolded': h_normalised_xsection_unfolded
}
histograms_normalised_xsection_systematics_shifts = {
'measured': h_normalised_xsection,
'unfolded': h_normalised_xsection_unfolded
}
if category == 'central':
# true distributions
h_normalised_xsection_MADGRAPH = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['MADGRAPH'], bin_edges[variable])
h_normalised_xsection_POWHEG_PYTHIA = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['POWHEG_PYTHIA'], bin_edges[variable])
h_normalised_xsection_POWHEG_HERWIG = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['POWHEG_HERWIG'], bin_edges[variable])
h_normalised_xsection_MCATNLO = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['MCATNLO'], bin_edges[variable])
h_normalised_xsection_mathchingup = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['matchingup'], bin_edges[variable])
h_normalised_xsection_mathchingdown = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['matchingdown'], bin_edges[variable])
h_normalised_xsection_scaleup = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['scaleup'], bin_edges[variable])
h_normalised_xsection_scaledown = value_error_tuplelist_to_hist(
normalised_xsection_unfolded['scaledown'], bin_edges[variable])
histograms_normalised_xsection_different_generators.update({
'MADGRAPH':
h_normalised_xsection_MADGRAPH,
'POWHEG_PYTHIA':
h_normalised_xsection_POWHEG_PYTHIA,
'POWHEG_HERWIG':
h_normalised_xsection_POWHEG_HERWIG,
'MCATNLO':
h_normalised_xsection_MCATNLO
})
histograms_normalised_xsection_systematics_shifts.update({
'MADGRAPH':
h_normalised_xsection_MADGRAPH,
'matchingdown':
h_normalised_xsection_mathchingdown,
'matchingup':
h_normalised_xsection_mathchingup,
'scaledown':
h_normalised_xsection_scaledown,
'scaleup':
h_normalised_xsection_scaleup
})
file_template = path_to_JSON + '/xsection_measurement_results/' + channel + \
'/kv' + str(k_values[channel]) + '/' + \
category + '/normalised_xsection_' + met_type
if channel == 'combined':
file_template = file_template.replace(
'kv' + str(k_values[channel]), '')
# normalised_xsection_unfolded_with_errors = read_data_from_JSON( file_template + '_with_errors.txt' )
normalised_xsection_unfolded_with_errors_with_systematics_but_without_ttbar_theory = read_data_from_JSON(
file_template +
'_with_systematics_but_without_ttbar_theory_errors.txt')
normalised_xsection_unfolded_with_errors_with_systematics_but_without_generator = read_data_from_JSON(
file_template +
'_with_systematics_but_without_generator_errors.txt')
# a rootpy.Graph with asymmetric errors!
h_normalised_xsection_with_systematics_but_without_ttbar_theory = value_errors_tuplelist_to_graph(
normalised_xsection_unfolded_with_errors_with_systematics_but_without_ttbar_theory[
'TTJet_measured'], bin_edges[variable])
h_normalised_xsection_with_systematics_but_without_ttbar_theory_unfolded = value_errors_tuplelist_to_graph(
normalised_xsection_unfolded_with_errors_with_systematics_but_without_ttbar_theory[
'TTJet_unfolded'], bin_edges[variable])
h_normalised_xsection_with_systematics_but_without_generator = value_errors_tuplelist_to_graph(
normalised_xsection_unfolded_with_errors_with_systematics_but_without_generator[
'TTJet_measured'], bin_edges[variable])
h_normalised_xsection_with_systematics_but_without_generator_unfolded = value_errors_tuplelist_to_graph(
normalised_xsection_unfolded_with_errors_with_systematics_but_without_generator[
'TTJet_unfolded'], bin_edges[variable])
histograms_normalised_xsection_different_generators[
'measured_with_systematics'] = h_normalised_xsection_with_systematics_but_without_generator
histograms_normalised_xsection_different_generators[
'unfolded_with_systematics'] = h_normalised_xsection_with_systematics_but_without_generator_unfolded
histograms_normalised_xsection_systematics_shifts[
'measured_with_systematics'] = h_normalised_xsection_with_systematics_but_without_ttbar_theory
histograms_normalised_xsection_systematics_shifts[
'unfolded_with_systematics'] = h_normalised_xsection_with_systematics_but_without_ttbar_theory_unfolded
return histograms_normalised_xsection_different_generators, histograms_normalised_xsection_systematics_shifts
