def plot_bias(h_unfold_model, h_data_model, unfolded_data, variable,
channel, come, method):
hp = Histogram_properties()
hp.name = '{channel}_bias_test_for_{variable}_at_{come}TeV'.format(
channel=channel,
variable=variable,
come=come,
)
v_latex = latex_labels.variables_latex[variable]
unit = ''
if variable in ['HT', 'ST', 'MET', 'WPT']:
unit = ' [GeV]'
hp.x_axis_title = v_latex + unit
hp.y_axis_title = 'Events'
hp.title = 'Closure tests for {variable}'.format(variable=v_latex)
output_folder = 'plots/unfolding/bias_test/{0}/'.format(method)
compare_measurements(models={'MC truth': h_data_model,
'unfold model': h_unfold_model},
measurements={'unfolded reco': unfolded_data},
show_measurement_errors=True,
histogram_properties=hp,
save_folder=output_folder,
save_as=['png', 'pdf'])
def plot_bias(h_unfold_model, h_data_model, unfolded_data, variable, channel,
come, method):
hp = Histogram_properties()
hp.name = '{channel}_bias_test_for_{variable}_at_{come}TeV'.format(
channel=channel,
variable=variable,
come=come,
)
v_latex = latex_labels.variables_latex[variable]
unit = ''
if variable in ['HT', 'ST', 'MET', 'WPT']:
unit = ' [GeV]'
hp.x_axis_title = v_latex + unit
hp.y_axis_title = 'Events'
hp.title = 'Closure tests for {variable}'.format(variable=v_latex)
output_folder = 'plots/unfolding/bias_test/{0}/'.format(method)
compare_measurements(models={
'MC truth': h_data_model,
'unfold model': h_unfold_model
},
measurements={'unfolded reco': unfolded_data},
show_measurement_errors=True,
histogram_properties=hp,
save_folder=output_folder,
save_as=['png', 'pdf'])
def plot_bias_in_all_bins(biases, mean_bias, centre_of_mass, channel, variable,
tau_value, output_folder, output_formats, bin_edges):
h_bias = Hist(bin_edges, type='D')
n_bins = h_bias.nbins()
assert len(biases) == n_bins
for i, bias in enumerate(biases):
h_bias.SetBinContent(i + 1, bias)
histogram_properties = Histogram_properties()
name_mpt = 'bias_{0}_{1}_{2}TeV'
histogram_properties.name = name_mpt.format(variable, channel,
centre_of_mass)
histogram_properties.y_axis_title = 'Bias'
histogram_properties.x_axis_title = latex_labels.variables_latex[variable]
title = 'pull distribution mean \& sigma for {0}'.format(tau_value)
histogram_properties.title = title
histogram_properties.y_limits = [0, 10]
histogram_properties.xerr = True
compare_measurements(
models={'Mean bias': make_line_hist(bin_edges, mean_bias)},
measurements={'Bias': h_bias},
show_measurement_errors=True,
histogram_properties=histogram_properties,
save_folder=output_folder,
save_as=output_formats)
def plot_bias_in_all_bins(biases, mean_bias, centre_of_mass, channel, variable,
tau_value, output_folder, output_formats, bin_edges):
h_bias = Hist(bin_edges, type='D')
n_bins = h_bias.nbins()
assert len(biases) == n_bins
for i, bias in enumerate(biases):
h_bias.SetBinContent(i+1, bias)
histogram_properties = Histogram_properties()
name_mpt = 'bias_{0}_{1}_{2}TeV'
histogram_properties.name = name_mpt.format(
variable,
channel,
centre_of_mass
)
histogram_properties.y_axis_title = 'Bias'
histogram_properties.x_axis_title = latex_labels.variables_latex[variable]
title = 'pull distribution mean \& sigma for {0}'.format(tau_value)
histogram_properties.title = title
histogram_properties.y_limits = [0, 10]
histogram_properties.xerr = True
compare_measurements(
models={
'Mean bias':make_line_hist(bin_edges, mean_bias)
},
measurements={
'Bias': h_bias
},
show_measurement_errors=True,
histogram_properties=histogram_properties,
save_folder=output_folder,
save_as=output_formats)
def plot_closure(unfolded_and_truths, variable, channel, come, method):
hp = Histogram_properties()
hp.name = '{channel}_closure_test_for_{variable}_at_{come}TeV'.format(
channel=channel,
variable=variable,
come=come,
)
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
# plt.ylabel( r, CMS.y_axis_title )
hp.y_axis_title = r'$\frac{1}{\sigma} \frac{d\sigma}{d' + v_latex + '}$' + unit
hp.title = 'Closure tests for {variable}'.format(variable=v_latex)
output_folder = 'plots/unfolding/closure_test/{0}/'.format(method)
models = OrderedDict()
measurements = OrderedDict()
for sample in unfolded_and_truths:
models[sample + ' truth'] = unfolded_and_truths[sample]['truth']
measurements[sample + ' unfolded'] = unfolded_and_truths[sample]['unfolded']
compare_measurements(
models = models,
measurements = measurements,
show_measurement_errors=True,
histogram_properties=hp,
save_folder=output_folder,
save_as=['pdf'],
match_models_to_measurements = True)
def plot_bias(unfolded_and_truths, variable, channel, come, method):
hp = Histogram_properties()
hp.name = 'Bias_{channel}_{variable}_at_{come}TeV'.format(
channel=channel,
variable=variable,
come=come,
)
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
# plt.ylabel( r, CMS.y_axis_title )
hp.y_axis_title = 'Unfolded / Truth'
hp.y_limits = [0.92, 1.08]
hp.title = 'Bias for {variable}'.format(variable=v_latex)
output_folder = 'plots/unfolding/bias_test/'
measurements = { 'Central' : unfolded_and_truths['Central']['bias'] }
models = {}
for sample in unfolded_and_truths:
if sample == 'Central' : continue
models[sample] = unfolded_and_truths[sample]['bias']
compare_measurements(
models = models,
measurements = measurements,
show_measurement_errors=True,
histogram_properties=hp,
save_folder=output_folder,
save_as=['pdf'],
match_models_to_measurements = True)
def plot_results(results):
'''
Takes results fo the form:
{centre-of-mass-energy: {
channel : {
variable : {
fit_variable : {
test : { sample : []},
}
}
}
}
}
'''
global options
output_base = 'plots/fit_checks/chi2'
for COMEnergy in results.keys():
tmp_result_1 = results[COMEnergy]
for channel in tmp_result_1.keys():
tmp_result_2 = tmp_result_1[channel]
for variable in tmp_result_2.keys():
tmp_result_3 = tmp_result_2[variable]
for fit_variable in tmp_result_3.keys():
tmp_result_4 = tmp_result_3[fit_variable]
# histograms should be {sample: {test : histogram}}
histograms = {}
for test, chi2 in tmp_result_4.iteritems():
for sample in chi2.keys():
if not histograms.has_key(sample):
histograms[sample] = {}
# reverse order of test and sample
histograms[sample][test] = value_tuplelist_to_hist(
chi2[sample], bin_edges_vis[variable])
for sample in histograms.keys():
hist_properties = Histogram_properties()
hist_properties.name = sample.replace('+',
'') + '_chi2'
hist_properties.title = '$\\chi^2$ distribution for fit output (' + sample + ')'
hist_properties.x_axis_title = '$' + latex_labels.variables_latex[
variable] + '$ [TeV]'
hist_properties.y_axis_title = '$\chi^2 = \\left({N_{fit}} - N_{{exp}}\\right)^2$'
hist_properties.set_log_y = True
hist_properties.y_limits = (1e-20, 1e20)
path = output_base + '/' + COMEnergy + 'TeV/' + channel + '/' + variable + '/' + fit_variable + '/'
if options.test:
path = output_base + '/test/'
measurements = {}
for test, histogram in histograms[sample].iteritems():
measurements[test.replace('_', ' ')] = histogram
compare_measurements(
{},
measurements,
show_measurement_errors=False,
histogram_properties=hist_properties,
save_folder=path,
save_as=['pdf'])
def compare_vjets_templates( variable = 'MET', met_type = 'patType1CorrectedPFMet',
title = 'Untitled', channel = 'electron' ):
''' Compares the V+jets templates in different bins
of the current variable'''
global fit_variable_properties, b_tag_bin, save_as
variable_bins = variable_bins_ROOT[variable]
histogram_template = get_histogram_template( variable )
for fit_variable in electron_fit_variables:
all_hists = {}
inclusive_hist = None
save_path = 'plots/%dTeV/fit_variables/%s/%s/' % ( measurement_config.centre_of_mass_energy, variable, fit_variable )
make_folder_if_not_exists( save_path + '/vjets/' )
max_bins = len( variable_bins )
for bin_range in variable_bins[0:max_bins]:
params = {'met_type': met_type, 'bin_range':bin_range, 'fit_variable':fit_variable, 'b_tag_bin':b_tag_bin, 'variable':variable}
fit_variable_distribution = histogram_template % params
# format: histograms['data'][qcd_fit_variable_distribution]
histograms = get_histograms_from_files( [fit_variable_distribution], histogram_files )
prepare_histograms( histograms, rebin = fit_variable_properties[fit_variable]['rebin'], scale_factor = measurement_config.luminosity_scale )
all_hists[bin_range] = histograms['V+Jets'][fit_variable_distribution]
# create the inclusive distributions
inclusive_hist = deepcopy( all_hists[variable_bins[0]] )
for bin_range in variable_bins[1:max_bins]:
inclusive_hist += all_hists[bin_range]
for bin_range in variable_bins[0:max_bins]:
if not all_hists[bin_range].Integral() == 0:
all_hists[bin_range].Scale( 1 / all_hists[bin_range].Integral() )
# normalise all histograms
inclusive_hist.Scale( 1 / inclusive_hist.Integral() )
# now compare inclusive to all bins
histogram_properties = Histogram_properties()
histogram_properties.x_axis_title = fit_variable_properties[fit_variable]['x-title']
histogram_properties.y_axis_title = fit_variable_properties[fit_variable]['y-title']
histogram_properties.y_axis_title = histogram_properties.y_axis_title.replace( 'Events', 'a.u.' )
histogram_properties.x_limits = [fit_variable_properties[fit_variable]['min'], fit_variable_properties[fit_variable]['max']]
histogram_properties.title = title
histogram_properties.additional_text = channel_latex[channel] + ', ' + b_tag_bins_latex[b_tag_bin]
histogram_properties.name = variable + '_' + fit_variable + '_' + b_tag_bin + '_VJets_template_comparison'
histogram_properties.y_max_scale = 1.5
measurements = {bin_range + ' GeV': histogram for bin_range, histogram in all_hists.iteritems()}
measurements = OrderedDict( sorted( measurements.items() ) )
fit_var = fit_variable.replace( 'electron_', '' )
fit_var = fit_var.replace( 'muon_', '' )
graphs = spread_x( measurements.values(), fit_variable_bin_edges[fit_var] )
for key, graph in zip( sorted( measurements.keys() ), graphs ):
measurements[key] = graph
compare_measurements( models = {'inclusive' : inclusive_hist},
measurements = measurements,
show_measurement_errors = True,
histogram_properties = histogram_properties,
save_folder = save_path + '/vjets/',
save_as = save_as )
def compare( central_mc, expected_result = None, measured_result = None, results = {}, variable = 'MET',
channel = 'electron', bin_edges = [] ):
global input_file, plot_location, ttbar_xsection, luminosity, centre_of_mass, method, test, log_plots
channel_label = ''
if channel == 'electron':
channel_label = 'e+jets, $\geq$4 jets'
elif channel == 'muon':
channel_label = '$\mu$+jets, $\geq$4 jets'
else:
channel_label = '$e, \mu$ + jets combined, $\geq$4 jets'
if test == 'data':
title_template = 'CMS Preliminary, $\mathcal{L} = %.1f$ fb$^{-1}$ at $\sqrt{s}$ = %d TeV \n %s'
title = title_template % ( luminosity / 1000., centre_of_mass, channel_label )
else:
title_template = 'CMS Simulation at $\sqrt{s}$ = %d TeV \n %s'
title = title_template % ( centre_of_mass, channel_label )
models = {latex_labels.measurements_latex['MADGRAPH'] : central_mc}
if expected_result and test == 'data':
models.update({'fitted data' : expected_result})
# scale central MC to lumi
nEvents = input_file.EventFilter.EventCounter.GetBinContent( 1 ) # number of processed events
lumiweight = ttbar_xsection * luminosity / nEvents
central_mc.Scale( lumiweight )
elif expected_result:
models.update({'expected' : expected_result})
if measured_result and test != 'data':
models.update({'measured' : measured_result})
measurements = collections.OrderedDict()
for key, value in results['k_value_results'].iteritems():
measurements['k = ' + str( key )] = value
# get some spread in x
graphs = spread_x( measurements.values(), bin_edges )
for key, graph in zip( measurements.keys(), graphs ):
measurements[key] = graph
histogram_properties = Histogram_properties()
histogram_properties.name = channel + '_' + variable + '_' + method + '_' + test
histogram_properties.title = title + ', ' + latex_labels.b_tag_bins_latex['2orMoreBtags']
histogram_properties.x_axis_title = '$' + latex_labels.variables_latex[variable] + '$'
histogram_properties.y_axis_title = r'Events'
# histogram_properties.y_limits = [0, 0.03]
histogram_properties.x_limits = [bin_edges[0], bin_edges[-1]]
if log_plots:
histogram_properties.set_log_y = True
histogram_properties.name += '_log'
compare_measurements( models, measurements, show_measurement_errors = True,
histogram_properties = histogram_properties,
save_folder = plot_location, save_as = ['pdf'] )
def plot_results ( results ):
'''
Takes results fo the form:
{centre-of-mass-energy: {
channel : {
variable : {
fit_variable : {
test : { sample : []},
}
}
}
}
}
'''
global options
output_base = 'plots/fit_checks/chi2'
for COMEnergy in results.keys():
tmp_result_1 = results[COMEnergy]
for channel in tmp_result_1.keys():
tmp_result_2 = tmp_result_1[channel]
for variable in tmp_result_2.keys():
tmp_result_3 = tmp_result_2[variable]
for fit_variable in tmp_result_3.keys():
tmp_result_4 = tmp_result_3[fit_variable]
# histograms should be {sample: {test : histogram}}
histograms = {}
for test, chi2 in tmp_result_4.iteritems():
for sample in chi2.keys():
if not histograms.has_key(sample):
histograms[sample] = {}
# reverse order of test and sample
histograms[sample][test] = value_tuplelist_to_hist(chi2[sample], bin_edges_vis[variable])
for sample in histograms.keys():
hist_properties = Histogram_properties()
hist_properties.name = sample.replace('+', '') + '_chi2'
hist_properties.title = '$\\chi^2$ distribution for fit output (' + sample + ')'
hist_properties.x_axis_title = '$' + latex_labels.variables_latex[variable] + '$ [TeV]'
hist_properties.y_axis_title = '$\chi^2 = \\left({N_{fit}} - N_{{exp}}\\right)^2$'
hist_properties.set_log_y = True
hist_properties.y_limits = (1e-20, 1e20)
path = output_base + '/' + COMEnergy + 'TeV/' + channel + '/' + variable + '/' + fit_variable + '/'
if options.test:
path = output_base + '/test/'
measurements = {}
for test, histogram in histograms[sample].iteritems():
measurements[test.replace('_',' ')] = histogram
compare_measurements({},
measurements,
show_measurement_errors = False,
histogram_properties = hist_properties,
save_folder = path,
save_as = ['pdf'])
def plot_bias(unfolded_and_truths, variable, channel, come, method, prefix, plot_systematics=False):
hp = Histogram_properties()
hp.name = 'Bias_{prefix}_{channel}_{variable}_at_{come}TeV'.format(
prefix=prefix,
channel=channel,
variable=variable,
come=come,
)
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
# plt.ylabel( r, CMS.y_axis_title )
hp.y_axis_title = 'Unfolded / Truth'
hp.y_limits = [0.7, 1.5]
hp.title = 'Bias for {variable}'.format(variable=v_latex)
hp.legend_location = (0.98, 0.95)
output_folder = 'plots/unfolding/bias_test/'
measurements = {}
# measurements = { 'Central' : unfolded_and_truths['Central']['bias'] }
# for bin in range(0, unfolded_and_truths['Central']['bias'].GetNbinsX() + 1 ):
# unfolded_and_truths['Central']['bias'].SetBinError(bin,0)
models = OrderedDict()
lineStyles = []
for sample in unfolded_and_truths:
if sample == 'Central' or sample == 'Nominal':
lineStyles.append('dashed')
else:
lineStyles.append('dotted')
models[sample] = unfolded_and_truths[sample]['bias']
if plot_systematics:
models['systematicsup'], models['systematicsdown'] = get_systematics(variable,channel,come,method)
lineStyles.append('solid')
lineStyles.append('solid')
compare_measurements(
models = models,
measurements = measurements,
show_measurement_errors=True,
histogram_properties=hp,
save_folder=output_folder,
save_as=['pdf'],
line_styles_for_models = lineStyles,
match_models_to_measurements = True
)
def plot_fit_results(fit_results, centre_of_mass, channel, variable, k_value,
tau_value, output_folder, output_formats, bin_edges):
h_mean = Hist(bin_edges, type='D')
h_sigma = Hist(bin_edges, type='D')
n_bins = h_mean.nbins()
assert len(fit_results) == n_bins
mean_abs_pull = 0
for i, fr in enumerate(fit_results):
mean_abs_pull += abs(fr.mean)
h_mean.SetBinContent(i + 1, fr.mean)
h_mean.SetBinError(i + 1, fr.meanError)
h_sigma.SetBinContent(i + 1, fr.sigma)
h_sigma.SetBinError(i + 1, fr.sigmaError)
mean_abs_pull /= n_bins
histogram_properties = Histogram_properties()
name_mpt = 'pull_distribution_mean_and_sigma_{0}_{1}_{2}TeV'
histogram_properties.name = name_mpt.format(
variable,
channel,
centre_of_mass
)
histogram_properties.y_axis_title = r'$\mu_{\text{pull}}$ ($\sigma_{\text{pull}}$)'
histogram_properties.x_axis_title = latex_labels.variables_latex[variable]
histogram_properties.legend_location = (0.98, 0.48)
value = get_value_title(k_value, tau_value)
title = 'pull distribution mean \& sigma for {0}'.format(tau_value)
histogram_properties.title = title
histogram_properties.y_limits = [-2, 2]
histogram_properties.xerr = True
compare_measurements(
models={
# 'mean $|\mu|$':make_line_hist(bin_edges,mean_abs_pull),
'ideal $\mu$': make_line_hist(bin_edges, 0),
'ideal $\sigma$': make_line_hist(bin_edges, 1),
},
measurements={
r'$\mu_{\text{pull}}$': h_mean,
r'$\sigma_{\text{pull}}$': h_sigma
},
show_measurement_errors=True,
histogram_properties=histogram_properties,
save_folder=output_folder,
save_as=output_formats)
def plot_bias(unfolded_and_truths, variable, channel, come, method):
hp = Histogram_properties()
hp.name = 'Bias_{channel}_{variable}_at_{come}TeV'.format(
channel=channel,
variable=variable,
come=come,
)
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
# plt.ylabel( r, CMS.y_axis_title )
hp.y_axis_title = 'Unfolded / Truth'
hp.y_limits = [0.85, 1.15]
hp.title = 'Bias for {variable}'.format(variable=v_latex)
output_folder = 'plots/unfolding/bias_test/'
measurements = { 'Central' : unfolded_and_truths['Central']['bias'] }
for bin in range(0, unfolded_and_truths['Central']['bias'].GetNbinsX() + 1 ):
unfolded_and_truths['Central']['bias'].SetBinError(bin,0)
# central_truth = unfolded_and_truths['Central']['truth']
# for label, hists in unfolded_and_truths.iteritems():
# truth = hists['truth']
# print label
# for i,j in zip( list(truth.y()), list(central_truth.y())) :
# print abs(1-i/j)*100
models = {}
for sample in unfolded_and_truths:
if sample == 'Central' : continue
models[sample] = unfolded_and_truths[sample]['bias']
compare_measurements(
models = models,
measurements = measurements,
show_measurement_errors=True,
histogram_properties=hp,
save_folder=output_folder,
save_as=['pdf'],
match_models_to_measurements = True)
def plot_fit_results(fit_results, centre_of_mass, channel, variable, k_value,
tau_value, output_folder, output_formats, bin_edges):
h_mean = Hist(bin_edges, type='D')
h_sigma = Hist(bin_edges, type='D')
n_bins = h_mean.nbins()
assert len(fit_results) == n_bins
mean_abs_pull = 0
for i, fr in enumerate(fit_results):
mean_abs_pull += abs(fr.mean)
h_mean.SetBinContent(i + 1, fr.mean)
h_mean.SetBinError(i + 1, fr.meanError)
h_sigma.SetBinContent(i + 1, fr.sigma)
h_sigma.SetBinError(i + 1, fr.sigmaError)
mean_abs_pull /= n_bins
histogram_properties = Histogram_properties()
name_mpt = 'pull_distribution_mean_and_sigma_{0}_{1}_{2}TeV'
histogram_properties.name = name_mpt.format(variable, channel,
centre_of_mass)
histogram_properties.y_axis_title = r'$\mu_{\text{pull}}$ ($\sigma_{\text{pull}}$)'
histogram_properties.x_axis_title = latex_labels.variables_latex[variable]
histogram_properties.legend_location = (0.98, 0.48)
value = get_value_title(k_value, tau_value)
title = 'pull distribution mean \& sigma for {0}'.format(tau_value)
histogram_properties.title = title
histogram_properties.y_limits = [-2, 2]
histogram_properties.xerr = True
compare_measurements(
models={
# 'mean $|\mu|$':make_line_hist(bin_edges,mean_abs_pull),
'ideal $\mu$': make_line_hist(bin_edges, 0),
'ideal $\sigma$': make_line_hist(bin_edges, 1),
},
measurements={
r'$\mu_{\text{pull}}$': h_mean,
r'$\sigma_{\text{pull}}$': h_sigma
},
show_measurement_errors=True,
histogram_properties=histogram_properties,
save_folder=output_folder,
save_as=output_formats)
def compare_vjets_btag_regions( variable = 'MET', met_type = 'patType1CorrectedPFMet',
title = 'Untitled', channel = 'electron' ):
''' Compares the V+Jets template in different b-tag bins'''
global fit_variable_properties, b_tag_bin, save_as, b_tag_bin_ctl
b_tag_bin_ctl = '0orMoreBtag'
variable_bins = variable_bins_ROOT[variable]
histogram_template = get_histogram_template( variable )
for fit_variable in electron_fit_variables:
if '_bl' in fit_variable:
b_tag_bin_ctl = '1orMoreBtag'
else:
b_tag_bin_ctl = '0orMoreBtag'
save_path = 'plots/%dTeV/fit_variables/%s/%s/' % ( measurement_config.centre_of_mass_energy, variable, fit_variable )
make_folder_if_not_exists( save_path + '/vjets/' )
histogram_properties = Histogram_properties()
histogram_properties.x_axis_title = fit_variable_properties[fit_variable]['x-title']
histogram_properties.y_axis_title = fit_variable_properties[fit_variable]['y-title']
histogram_properties.y_axis_title = histogram_properties.y_axis_title.replace( 'Events', 'a.u.' )
histogram_properties.x_limits = [fit_variable_properties[fit_variable]['min'], fit_variable_properties[fit_variable]['max']]
histogram_properties.title = title
histogram_properties.additional_text = channel_latex[channel] + ', ' + b_tag_bins_latex[b_tag_bin_ctl]
histogram_properties.y_max_scale = 1.5
for bin_range in variable_bins:
params = {'met_type': met_type, 'bin_range':bin_range, 'fit_variable':fit_variable, 'b_tag_bin':b_tag_bin, 'variable':variable}
fit_variable_distribution = histogram_template % params
fit_variable_distribution_ctl = fit_variable_distribution.replace( b_tag_bin, b_tag_bin_ctl )
# format: histograms['data'][qcd_fit_variable_distribution]
histograms = get_histograms_from_files( [fit_variable_distribution, fit_variable_distribution_ctl], {'V+Jets' : histogram_files['V+Jets']} )
prepare_histograms( histograms, rebin = fit_variable_properties[fit_variable]['rebin'], scale_factor = measurement_config.luminosity_scale )
histogram_properties.name = variable + '_' + bin_range + '_' + fit_variable + '_' + b_tag_bin_ctl + '_VJets_template_comparison'
histograms['V+Jets'][fit_variable_distribution].Scale( 1 / histograms['V+Jets'][fit_variable_distribution].Integral() )
histograms['V+Jets'][fit_variable_distribution_ctl].Scale( 1 / histograms['V+Jets'][fit_variable_distribution_ctl].Integral() )
compare_measurements( models = {'no b-tag' : histograms['V+Jets'][fit_variable_distribution_ctl]},
measurements = {'$>=$ 2 b-tags': histograms['V+Jets'][fit_variable_distribution]},
show_measurement_errors = True,
histogram_properties = histogram_properties,
save_folder = save_path + '/vjets/',
save_as = save_as )
def plot_closure(unfolded_and_truths, variable, channel, come, method, quantity):
hp = Histogram_properties()
hp.name = '{quantity}_{channel}_closure_test_for_{variable}_at_{come}TeV'.format(
quantity=quantity,
channel=channel,
variable=variable,
come=come,
)
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
if quantity == 'number_of_unfolded_events':
hp.y_axis_title = 'Number of unfolded events'
elif quantity == 'normalised_xsection':
hp.y_axis_title = 'Normalised Cross Section'
hp.title = 'Closure tests for {variable}'.format(variable=v_latex)
output_folder = 'plots/unfolding/closure_test/'
models = OrderedDict()
measurements = OrderedDict()
for sample in unfolded_and_truths:
models[sample + ' truth'] = unfolded_and_truths[sample]['truth']
measurements[sample + ' unfolded'] = unfolded_and_truths[sample]['unfolded']
compare_measurements(
models = models,
measurements = measurements,
show_measurement_errors=True,
histogram_properties=hp,
save_folder=output_folder,
save_as=['pdf'],
match_models_to_measurements = True
)
def compare_vjets_templates(variable='MET',
met_type='patType1CorrectedPFMet',
title='Untitled',
channel='electron'):
''' Compares the V+jets templates in different bins
of the current variable'''
global fit_variable_properties, b_tag_bin, save_as
variable_bins = variable_bins_ROOT[variable]
histogram_template = get_histogram_template(variable)
for fit_variable in electron_fit_variables:
all_hists = {}
inclusive_hist = None
save_path = 'plots/%dTeV/fit_variables/%s/%s/' % (
measurement_config.centre_of_mass_energy, variable, fit_variable)
make_folder_if_not_exists(save_path + '/vjets/')
max_bins = len(variable_bins)
for bin_range in variable_bins[0:max_bins]:
params = {
'met_type': met_type,
'bin_range': bin_range,
'fit_variable': fit_variable,
'b_tag_bin': b_tag_bin,
'variable': variable
}
fit_variable_distribution = histogram_template % params
# format: histograms['data'][qcd_fit_variable_distribution]
histograms = get_histograms_from_files([fit_variable_distribution],
histogram_files)
prepare_histograms(
histograms,
rebin=fit_variable_properties[fit_variable]['rebin'],
scale_factor=measurement_config.luminosity_scale)
all_hists[bin_range] = histograms['V+Jets'][
fit_variable_distribution]
# create the inclusive distributions
inclusive_hist = deepcopy(all_hists[variable_bins[0]])
for bin_range in variable_bins[1:max_bins]:
inclusive_hist += all_hists[bin_range]
for bin_range in variable_bins[0:max_bins]:
if not all_hists[bin_range].Integral() == 0:
all_hists[bin_range].Scale(1 / all_hists[bin_range].Integral())
# normalise all histograms
inclusive_hist.Scale(1 / inclusive_hist.Integral())
# now compare inclusive to all bins
histogram_properties = Histogram_properties()
histogram_properties.x_axis_title = fit_variable_properties[
fit_variable]['x-title']
histogram_properties.y_axis_title = fit_variable_properties[
fit_variable]['y-title']
histogram_properties.y_axis_title = histogram_properties.y_axis_title.replace(
'Events', 'a.u.')
histogram_properties.x_limits = [
fit_variable_properties[fit_variable]['min'],
fit_variable_properties[fit_variable]['max']
]
histogram_properties.title = title
histogram_properties.additional_text = channel_latex[
channel] + ', ' + b_tag_bins_latex[b_tag_bin]
histogram_properties.name = variable + '_' + fit_variable + '_' + b_tag_bin + '_VJets_template_comparison'
histogram_properties.y_max_scale = 1.5
measurements = {
bin_range + ' GeV': histogram
for bin_range, histogram in all_hists.iteritems()
}
measurements = OrderedDict(sorted(measurements.items()))
fit_var = fit_variable.replace('electron_', '')
fit_var = fit_var.replace('muon_', '')
graphs = spread_x(measurements.values(),
fit_variable_bin_edges[fit_var])
for key, graph in zip(sorted(measurements.keys()), graphs):
measurements[key] = graph
compare_measurements(models={'inclusive': inclusive_hist},
measurements=measurements,
show_measurement_errors=True,
histogram_properties=histogram_properties,
save_folder=save_path + '/vjets/',
save_as=save_as)
def main():
config = XSectionConfig(13)
file_for_powhegPythia = File(config.unfolding_central_firstHalf, 'read')
file_for_ptReweight_up = File(config.unfolding_ptreweight_up_firstHalf,
'read')
file_for_ptReweight_down = File(config.unfolding_ptreweight_down_firstHalf,
'read')
file_for_amcatnlo_pythia8 = File(config.unfolding_amcatnlo_pythia8, 'read')
file_for_powhegHerwig = File(config.unfolding_powheg_herwig, 'read')
file_for_etaReweight_up = File(config.unfolding_etareweight_up, 'read')
file_for_etaReweight_down = File(config.unfolding_etareweight_down, 'read')
file_for_data_template = 'data/normalisation/background_subtraction/13TeV/{variable}/VisiblePS/central/normalisation_{channel}.txt'
for channel in config.analysis_types.keys():
if channel is 'combined': continue
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_pt_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_pt_reweighted_up = asrootpy(
response_pt_reweighted_up.ProjectionX('px', 1))
truth_pt_reweighted_up = asrootpy(
response_pt_reweighted_up.ProjectionY())
_, _, response_pt_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_pt_reweighted_down = asrootpy(
response_pt_reweighted_down.ProjectionX('px', 1))
truth_pt_reweighted_down = asrootpy(
response_pt_reweighted_down.ProjectionY())
# _, _, response_eta_reweighted_up, _ = get_unfold_histogram_tuple(
# inputfile=file_for_etaReweight_up,
# variable=variable,
# channel=channel,
# centre_of_mass=13,
# load_fakes=False,
# visiblePS=True
# )
# measured_eta_reweighted_up = asrootpy(response_eta_reweighted_up.ProjectionX('px',1))
# truth_eta_reweighted_up = asrootpy(response_eta_reweighted_up.ProjectionY())
# _, _, response_eta_reweighted_down, _ = get_unfold_histogram_tuple(
# inputfile=file_for_etaReweight_down,
# variable=variable,
# channel=channel,
# centre_of_mass=13,
# load_fakes=False,
# visiblePS=True
# )
# measured_eta_reweighted_down = asrootpy(response_eta_reweighted_down.ProjectionX('px',1))
# truth_eta_reweighted_down = asrootpy(response_eta_reweighted_down.ProjectionY())
# Get the distributions for other MC models
_, _, response_amcatnlo_pythia8, _ = get_unfold_histogram_tuple(
inputfile=file_for_amcatnlo_pythia8,
variable=variable,
channel=channel,
centre_of_mass=13,
load_fakes=False,
visiblePS=True)
measured_amcatnlo_pythia8 = asrootpy(
response_amcatnlo_pythia8.ProjectionX('px', 1))
truth_amcatnlo_pythia8 = asrootpy(
response_amcatnlo_pythia8.ProjectionY())
_, _, response_powhegHerwig, _ = get_unfold_histogram_tuple(
inputfile=file_for_powhegHerwig,
variable=variable,
channel=channel,
centre_of_mass=13,
load_fakes=False,
visiblePS=True)
measured_powhegHerwig = asrootpy(
response_powhegHerwig.ProjectionX('px', 1))
truth_powhegHerwig = asrootpy(response_powhegHerwig.ProjectionY())
# Get the data input (data after background subtraction, and fake removal)
file_for_data = file_for_data_template.format(variable=variable,
channel=channel)
data = read_tuple_from_file(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.x_limits = [
reco_bin_edges_vis[variable][0],
reco_bin_edges_vis[variable][-1]
]
hp.ratio_y_limits = [0.1, 1.9]
hp.ratio_y_title = 'Reweighted / Central'
hp.y_axis_title = 'Number of events'
hp.title = 'Reweighting check for {variable}'.format(
variable=v_latex)
measured_central.Rebin(2)
measured_pt_reweighted_up.Rebin(2)
measured_pt_reweighted_down.Rebin(2)
# measured_eta_reweighted_up.Rebin(2)
# measured_eta_reweighted_down.Rebin(2)
measured_amcatnlo_pythia8.Rebin(2)
measured_powhegHerwig.Rebin(2)
data.Rebin(2)
measured_central.Scale(1 / measured_central.Integral())
measured_pt_reweighted_up.Scale(
1 / measured_pt_reweighted_up.Integral())
measured_pt_reweighted_down.Scale(
1 / measured_pt_reweighted_down.Integral())
measured_amcatnlo_pythia8.Scale(
1 / measured_amcatnlo_pythia8.Integral())
measured_powhegHerwig.Scale(1 / measured_powhegHerwig.Integral())
# measured_eta_reweighted_up.Scale( 1 / measured_eta_reweighted_up.Integral() )
# measured_eta_reweighted_down.Scale( 1/ measured_eta_reweighted_down.Integral() )
data.Scale(1 / data.Integral())
print list(measured_central.y())
print list(measured_amcatnlo_pythia8.y())
print list(measured_powhegHerwig.y())
print list(data.y())
compare_measurements(
# models = {'Central' : measured_central, 'PtReweighted Up' : measured_pt_reweighted_up, 'PtReweighted Down' : measured_pt_reweighted_down, 'EtaReweighted Up' : measured_eta_reweighted_up, 'EtaReweighted Down' : measured_eta_reweighted_down},
models=OrderedDict([
('Central', measured_central),
('PtReweighted Up', measured_pt_reweighted_up),
('PtReweighted Down', measured_pt_reweighted_down),
('amc@nlo', measured_amcatnlo_pythia8),
('powhegHerwig', measured_powhegHerwig)
]),
measurements={'Data': data},
show_measurement_errors=True,
histogram_properties=hp,
save_folder='plots/unfolding/reweighting_check',
save_as=['pdf'],
line_styles_for_models=[
'solid', 'solid', 'solid', 'dashed', 'dashed'
],
show_ratio_for_pairs=OrderedDict([
('PtUpVsCentral',
[measured_pt_reweighted_up, measured_central]),
('PtDownVsCentral',
[measured_pt_reweighted_down, measured_central]),
('amcatnloVsCentral',
[measured_amcatnlo_pythia8, measured_central]),
('powhegHerwigVsCentral',
[measured_powhegHerwig, measured_central]),
('DataVsCentral', [data, measured_central])
]),
)
def main():
config = XSectionConfig(13)
file_for_powhegPythia = File(config.unfolding_central_firstHalf, 'read')
file_for_ptReweight_up = File(config.unfolding_ptreweight_up_firstHalf, 'read')
file_for_ptReweight_down = File(config.unfolding_ptreweight_down_firstHalf, 'read')
file_for_amcatnlo_pythia8 = File(config.unfolding_amcatnlo_pythia8, 'read')
file_for_powhegHerwig = File(config.unfolding_powheg_herwig, 'read')
file_for_etaReweight_up = File(config.unfolding_etareweight_up, 'read')
file_for_etaReweight_down = File(config.unfolding_etareweight_down, 'read')
file_for_data_template = 'data/normalisation/background_subtraction/13TeV/{variable}/VisiblePS/central/normalisation_{channel}.txt'
for channel in config.analysis_types.keys():
if channel is 'combined':continue
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_pt_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_pt_reweighted_up = asrootpy(response_pt_reweighted_up.ProjectionX('px',1))
truth_pt_reweighted_up = asrootpy(response_pt_reweighted_up.ProjectionY())
_, _, response_pt_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_pt_reweighted_down = asrootpy(response_pt_reweighted_down.ProjectionX('px',1))
truth_pt_reweighted_down = asrootpy(response_pt_reweighted_down.ProjectionY())
# _, _, response_eta_reweighted_up, _ = get_unfold_histogram_tuple(
# inputfile=file_for_etaReweight_up,
# variable=variable,
# channel=channel,
# centre_of_mass=13,
# load_fakes=False,
# visiblePS=True
# )
# measured_eta_reweighted_up = asrootpy(response_eta_reweighted_up.ProjectionX('px',1))
# truth_eta_reweighted_up = asrootpy(response_eta_reweighted_up.ProjectionY())
# _, _, response_eta_reweighted_down, _ = get_unfold_histogram_tuple(
# inputfile=file_for_etaReweight_down,
# variable=variable,
# channel=channel,
# centre_of_mass=13,
# load_fakes=False,
# visiblePS=True
# )
# measured_eta_reweighted_down = asrootpy(response_eta_reweighted_down.ProjectionX('px',1))
# truth_eta_reweighted_down = asrootpy(response_eta_reweighted_down.ProjectionY())
# Get the distributions for other MC models
_, _, response_amcatnlo_pythia8, _ = get_unfold_histogram_tuple(
inputfile=file_for_amcatnlo_pythia8,
variable=variable,
channel=channel,
centre_of_mass=13,
load_fakes=False,
visiblePS=True
)
measured_amcatnlo_pythia8 = asrootpy(response_amcatnlo_pythia8.ProjectionX('px',1))
truth_amcatnlo_pythia8 = asrootpy(response_amcatnlo_pythia8.ProjectionY())
_, _, response_powhegHerwig, _ = get_unfold_histogram_tuple(
inputfile=file_for_powhegHerwig,
variable=variable,
channel=channel,
centre_of_mass=13,
load_fakes=False,
visiblePS=True
)
measured_powhegHerwig = asrootpy(response_powhegHerwig.ProjectionX('px',1))
truth_powhegHerwig = asrootpy(response_powhegHerwig.ProjectionY())
# Get the data input (data after background subtraction, and fake removal)
file_for_data = file_for_data_template.format( variable = variable, channel = channel )
data = read_tuple_from_file(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.x_limits = [ reco_bin_edges_vis[variable][0], reco_bin_edges_vis[variable][-1]]
hp.ratio_y_limits = [0.1,1.9]
hp.ratio_y_title = 'Reweighted / Central'
hp.y_axis_title = 'Number of events'
hp.title = 'Reweighting check for {variable}'.format(variable=v_latex)
measured_central.Rebin(2)
measured_pt_reweighted_up.Rebin(2)
measured_pt_reweighted_down.Rebin(2)
# measured_eta_reweighted_up.Rebin(2)
# measured_eta_reweighted_down.Rebin(2)
measured_amcatnlo_pythia8.Rebin(2)
measured_powhegHerwig.Rebin(2)
data.Rebin(2)
measured_central.Scale( 1 / measured_central.Integral() )
measured_pt_reweighted_up.Scale( 1 / measured_pt_reweighted_up.Integral() )
measured_pt_reweighted_down.Scale( 1 / measured_pt_reweighted_down.Integral() )
measured_amcatnlo_pythia8.Scale( 1 / measured_amcatnlo_pythia8.Integral() )
measured_powhegHerwig.Scale( 1 / measured_powhegHerwig.Integral() )
# measured_eta_reweighted_up.Scale( 1 / measured_eta_reweighted_up.Integral() )
# measured_eta_reweighted_down.Scale( 1/ measured_eta_reweighted_down.Integral() )
data.Scale( 1 / data.Integral() )
print list(measured_central.y())
print list(measured_amcatnlo_pythia8.y())
print list(measured_powhegHerwig.y())
print list(data.y())
compare_measurements(
# models = {'Central' : measured_central, 'PtReweighted Up' : measured_pt_reweighted_up, 'PtReweighted Down' : measured_pt_reweighted_down, 'EtaReweighted Up' : measured_eta_reweighted_up, 'EtaReweighted Down' : measured_eta_reweighted_down},
models = OrderedDict([('Central' , measured_central), ('PtReweighted Up' , measured_pt_reweighted_up), ('PtReweighted Down' , measured_pt_reweighted_down), ('amc@nlo' , measured_amcatnlo_pythia8), ('powhegHerwig' , measured_powhegHerwig) ] ),
measurements = {'Data' : data},
show_measurement_errors=True,
histogram_properties=hp,
save_folder='plots/unfolding/reweighting_check',
save_as=['pdf'],
line_styles_for_models = ['solid','solid','solid','dashed','dashed'],
show_ratio_for_pairs = OrderedDict( [
('PtUpVsCentral' , [ measured_pt_reweighted_up, measured_central ] ),
('PtDownVsCentral' , [ measured_pt_reweighted_down, measured_central ] ),
('amcatnloVsCentral' , [ measured_amcatnlo_pythia8, measured_central ] ),
('powhegHerwigVsCentral' , [ measured_powhegHerwig, measured_central ] ),
('DataVsCentral' , [data, measured_central] )
]),
)
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 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_etaReweight_up = File(config.unfolding_etareweight_up, "read")
file_for_etaReweight_down = File(config.unfolding_etareweight_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_pt_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_pt_reweighted_up = asrootpy(response_pt_reweighted_up.ProjectionX("px", 1))
truth_pt_reweighted_up = asrootpy(response_pt_reweighted_up.ProjectionY())
_, _, response_pt_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_pt_reweighted_down = asrootpy(response_pt_reweighted_down.ProjectionX("px", 1))
truth_pt_reweighted_down = asrootpy(response_pt_reweighted_down.ProjectionY())
_, _, response_eta_reweighted_up, _ = get_unfold_histogram_tuple(
inputfile=file_for_etaReweight_up,
variable=variable,
channel=channel,
centre_of_mass=13,
load_fakes=False,
visiblePS=True,
)
measured_eta_reweighted_up = asrootpy(response_eta_reweighted_up.ProjectionX("px", 1))
truth_eta_reweighted_up = asrootpy(response_eta_reweighted_up.ProjectionY())
_, _, response_eta_reweighted_down, _ = get_unfold_histogram_tuple(
inputfile=file_for_etaReweight_down,
variable=variable,
channel=channel,
centre_of_mass=13,
load_fakes=False,
visiblePS=True,
)
measured_eta_reweighted_down = asrootpy(response_eta_reweighted_down.ProjectionX("px", 1))
truth_eta_reweighted_down = asrootpy(response_eta_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_pt_reweighted_up.Rebin(2)
measured_pt_reweighted_down.Rebin(2)
measured_eta_reweighted_up.Rebin(2)
measured_eta_reweighted_down.Rebin(2)
data.Rebin(2)
measured_central.Scale(1 / measured_central.Integral())
measured_pt_reweighted_up.Scale(1 / measured_pt_reweighted_up.Integral())
measured_pt_reweighted_down.Scale(1 / measured_pt_reweighted_down.Integral())
measured_eta_reweighted_up.Scale(1 / measured_eta_reweighted_up.Integral())
measured_eta_reweighted_down.Scale(1 / measured_eta_reweighted_down.Integral())
data.Scale(1 / data.Integral())
compare_measurements(
models={
"Central": measured_central,
"PtReweighted Up": measured_pt_reweighted_up,
"PtReweighted Down": measured_pt_reweighted_down,
"EtaReweighted Up": measured_eta_reweighted_up,
"EtaReweighted Down": measured_eta_reweighted_down,
},
measurements={"Data": data},
show_measurement_errors=True,
histogram_properties=hp,
save_folder="plots/unfolding/reweighting_check",
save_as=["pdf"],
)
def compare_qcd_control_regions( variable = 'MET', met_type = 'patType1CorrectedPFMet', title = 'Untitled', channel = 'electron' ):
''' Compares the templates from the control regions in different bins
of the current variable'''
global fit_variable_properties, b_tag_bin, save_as, b_tag_bin_ctl
variable_bins = variable_bins_ROOT[variable]
histogram_template = get_histogram_template( variable )
for fit_variable in electron_fit_variables:
all_hists = {}
inclusive_hist = None
if '_bl' in fit_variable:
b_tag_bin_ctl = '1orMoreBtag'
else:
b_tag_bin_ctl = '0orMoreBtag'
save_path = 'plots/%dTeV/fit_variables/%s/%s/' % ( measurement_config.centre_of_mass_energy, variable, fit_variable )
make_folder_if_not_exists( save_path + '/qcd/' )
max_bins = 3
for bin_range in variable_bins[0:max_bins]:
params = {'met_type': met_type, 'bin_range':bin_range, 'fit_variable':fit_variable, 'b_tag_bin':b_tag_bin, 'variable':variable}
fit_variable_distribution = histogram_template % params
qcd_fit_variable_distribution = fit_variable_distribution.replace( 'Ref selection', 'QCDConversions' )
qcd_fit_variable_distribution = qcd_fit_variable_distribution.replace( b_tag_bin, b_tag_bin_ctl )
# format: histograms['data'][qcd_fit_variable_distribution]
histograms = get_histograms_from_files( [qcd_fit_variable_distribution], histogram_files )
prepare_histograms( histograms, rebin = fit_variable_properties[fit_variable]['rebin'], scale_factor = measurement_config.luminosity_scale )
histograms_for_cleaning = {'data':histograms['data'][qcd_fit_variable_distribution],
'V+Jets':histograms['V+Jets'][qcd_fit_variable_distribution],
'SingleTop':histograms['SingleTop'][qcd_fit_variable_distribution],
'TTJet':histograms['TTJet'][qcd_fit_variable_distribution]}
qcd_from_data = clean_control_region( histograms_for_cleaning, subtract = ['TTJet', 'V+Jets', 'SingleTop'] )
# clean
all_hists[bin_range] = qcd_from_data
# create the inclusive distributions
inclusive_hist = deepcopy( all_hists[variable_bins[0]] )
for bin_range in variable_bins[1:max_bins]:
inclusive_hist += all_hists[bin_range]
for bin_range in variable_bins[0:max_bins]:
if not all_hists[bin_range].Integral() == 0:
all_hists[bin_range].Scale( 1 / all_hists[bin_range].Integral() )
# normalise all histograms
inclusive_hist.Scale( 1 / inclusive_hist.Integral() )
# now compare inclusive to all bins
histogram_properties = Histogram_properties()
histogram_properties.x_axis_title = fit_variable_properties[fit_variable]['x-title']
histogram_properties.y_axis_title = fit_variable_properties[fit_variable]['y-title']
histogram_properties.y_axis_title = histogram_properties.y_axis_title.replace( 'Events', 'a.u.' )
histogram_properties.x_limits = [fit_variable_properties[fit_variable]['min'], fit_variable_properties[fit_variable]['max']]
# histogram_properties.y_limits = [0, 0.5]
histogram_properties.title = title
histogram_properties.additional_text = channel_latex[channel] + ', ' + b_tag_bins_latex[b_tag_bin_ctl]
histogram_properties.name = variable + '_' + fit_variable + '_' + b_tag_bin_ctl + '_QCD_template_comparison'
histogram_properties.y_max_scale = 1.5
measurements = {bin_range + ' GeV': histogram for bin_range, histogram in all_hists.iteritems()}
measurements = OrderedDict( sorted( measurements.items() ) )
compare_measurements( models = {'inclusive' : inclusive_hist},
measurements = measurements,
show_measurement_errors = True,
histogram_properties = histogram_properties,
save_folder = save_path + '/qcd/',
save_as = save_as )
def compare_vjets_btag_regions(variable='MET',
met_type='patType1CorrectedPFMet',
title='Untitled',
channel='electron'):
''' Compares the V+Jets template in different b-tag bins'''
global fit_variable_properties, b_tag_bin, save_as, b_tag_bin_ctl
b_tag_bin_ctl = '0orMoreBtag'
variable_bins = variable_bins_ROOT[variable]
histogram_template = get_histogram_template(variable)
for fit_variable in electron_fit_variables:
if '_bl' in fit_variable:
b_tag_bin_ctl = '1orMoreBtag'
else:
b_tag_bin_ctl = '0orMoreBtag'
save_path = 'plots/%dTeV/fit_variables/%s/%s/' % (
measurement_config.centre_of_mass_energy, variable, fit_variable)
make_folder_if_not_exists(save_path + '/vjets/')
histogram_properties = Histogram_properties()
histogram_properties.x_axis_title = fit_variable_properties[
fit_variable]['x-title']
histogram_properties.y_axis_title = fit_variable_properties[
fit_variable]['y-title']
histogram_properties.y_axis_title = histogram_properties.y_axis_title.replace(
'Events', 'a.u.')
histogram_properties.x_limits = [
fit_variable_properties[fit_variable]['min'],
fit_variable_properties[fit_variable]['max']
]
histogram_properties.title = title
histogram_properties.additional_text = channel_latex[
channel] + ', ' + b_tag_bins_latex[b_tag_bin_ctl]
histogram_properties.y_max_scale = 1.5
for bin_range in variable_bins:
params = {
'met_type': met_type,
'bin_range': bin_range,
'fit_variable': fit_variable,
'b_tag_bin': b_tag_bin,
'variable': variable
}
fit_variable_distribution = histogram_template % params
fit_variable_distribution_ctl = fit_variable_distribution.replace(
b_tag_bin, b_tag_bin_ctl)
# format: histograms['data'][qcd_fit_variable_distribution]
histograms = get_histograms_from_files(
[fit_variable_distribution, fit_variable_distribution_ctl],
{'V+Jets': histogram_files['V+Jets']})
prepare_histograms(
histograms,
rebin=fit_variable_properties[fit_variable]['rebin'],
scale_factor=measurement_config.luminosity_scale)
histogram_properties.name = variable + '_' + bin_range + '_' + fit_variable + '_' + b_tag_bin_ctl + '_VJets_template_comparison'
histograms['V+Jets'][fit_variable_distribution].Scale(
1 / histograms['V+Jets'][fit_variable_distribution].Integral())
histograms['V+Jets'][fit_variable_distribution_ctl].Scale(
1 /
histograms['V+Jets'][fit_variable_distribution_ctl].Integral())
compare_measurements(
models={
'no b-tag':
histograms['V+Jets'][fit_variable_distribution_ctl]
},
measurements={
'$>=$ 2 b-tags':
histograms['V+Jets'][fit_variable_distribution]
},
show_measurement_errors=True,
histogram_properties=histogram_properties,
save_folder=save_path + '/vjets/',
save_as=save_as)
def compare_qcd_control_regions(variable='MET',
met_type='patType1CorrectedPFMet',
title='Untitled',
channel='electron'):
''' Compares the templates from the control regions in different bins
of the current variable'''
global fit_variable_properties, b_tag_bin, save_as, b_tag_bin_ctl
variable_bins = variable_bins_ROOT[variable]
histogram_template = get_histogram_template(variable)
for fit_variable in electron_fit_variables:
all_hists = {}
inclusive_hist = None
if '_bl' in fit_variable:
b_tag_bin_ctl = '1orMoreBtag'
else:
b_tag_bin_ctl = '0orMoreBtag'
save_path = 'plots/%dTeV/fit_variables/%s/%s/' % (
measurement_config.centre_of_mass_energy, variable, fit_variable)
make_folder_if_not_exists(save_path + '/qcd/')
max_bins = 3
for bin_range in variable_bins[0:max_bins]:
params = {
'met_type': met_type,
'bin_range': bin_range,
'fit_variable': fit_variable,
'b_tag_bin': b_tag_bin,
'variable': variable
}
fit_variable_distribution = histogram_template % params
qcd_fit_variable_distribution = fit_variable_distribution.replace(
'Ref selection', 'QCDConversions')
qcd_fit_variable_distribution = qcd_fit_variable_distribution.replace(
b_tag_bin, b_tag_bin_ctl)
# format: histograms['data'][qcd_fit_variable_distribution]
histograms = get_histograms_from_files(
[qcd_fit_variable_distribution], histogram_files)
prepare_histograms(
histograms,
rebin=fit_variable_properties[fit_variable]['rebin'],
scale_factor=measurement_config.luminosity_scale)
histograms_for_cleaning = {
'data': histograms['data'][qcd_fit_variable_distribution],
'V+Jets': histograms['V+Jets'][qcd_fit_variable_distribution],
'SingleTop':
histograms['SingleTop'][qcd_fit_variable_distribution],
'TTJet': histograms['TTJet'][qcd_fit_variable_distribution]
}
qcd_from_data = clean_control_region(
histograms_for_cleaning,
subtract=['TTJet', 'V+Jets', 'SingleTop'])
# clean
all_hists[bin_range] = qcd_from_data
# create the inclusive distributions
inclusive_hist = deepcopy(all_hists[variable_bins[0]])
for bin_range in variable_bins[1:max_bins]:
inclusive_hist += all_hists[bin_range]
for bin_range in variable_bins[0:max_bins]:
if not all_hists[bin_range].Integral() == 0:
all_hists[bin_range].Scale(1 / all_hists[bin_range].Integral())
# normalise all histograms
inclusive_hist.Scale(1 / inclusive_hist.Integral())
# now compare inclusive to all bins
histogram_properties = Histogram_properties()
histogram_properties.x_axis_title = fit_variable_properties[
fit_variable]['x-title']
histogram_properties.y_axis_title = fit_variable_properties[
fit_variable]['y-title']
histogram_properties.y_axis_title = histogram_properties.y_axis_title.replace(
'Events', 'a.u.')
histogram_properties.x_limits = [
fit_variable_properties[fit_variable]['min'],
fit_variable_properties[fit_variable]['max']
]
# histogram_properties.y_limits = [0, 0.5]
histogram_properties.title = title
histogram_properties.additional_text = channel_latex[
channel] + ', ' + b_tag_bins_latex[b_tag_bin_ctl]
histogram_properties.name = variable + '_' + fit_variable + '_' + b_tag_bin_ctl + '_QCD_template_comparison'
histogram_properties.y_max_scale = 1.5
measurements = {
bin_range + ' GeV': histogram
for bin_range, histogram in all_hists.iteritems()
}
measurements = OrderedDict(sorted(measurements.items()))
compare_measurements(models={'inclusive': inclusive_hist},
measurements=measurements,
show_measurement_errors=True,
histogram_properties=histogram_properties,
save_folder=save_path + '/qcd/',
save_as=save_as)
