def calculate_normalisation(self):
'''
1. get file names
2. get histograms from files
3. ???
4. calculate normalisation based on self.method
'''
if self.have_normalisation:
return
histograms = self.measurement.histograms
for sample, hist in histograms.items():
# TODO: this should be a list of bin-contents
hist = fix_overflow(hist)
histograms[sample] = hist
self.initial_normalisation[
sample] = hist_to_value_error_tuplelist(hist)
if self.method == self.BACKGROUND_SUBTRACTION and sample != 'TTJet':
self.normalisation[sample] = self.initial_normalisation[sample]
if self.method == self.BACKGROUND_SUBTRACTION:
self.background_subtraction(histograms)
if self.method == self.SIMULTANEOUS_FIT:
self.simultaneous_fit(histograms)
# next, let's round all numbers (they are event numbers after all
for sample, values in self.normalisation.items():
new_values = [(round(v, 1), round(e, 1)) for v, e in values]
self.normalisation[sample] = new_values
self.have_normalisation = True
def calculate_normalisation(self):
'''
1. get file names
2. get histograms from files
3. ???
4. calculate normalisation
'''
if self.have_normalisation:
return
histograms = self.measurement.histograms
for sample, hist in histograms.items():
# TODO: this should be a list of bin-contents
hist = fix_overflow(hist)
histograms[sample] = hist
self.initial_normalisation[sample] = hist_to_value_error_tuplelist(
hist)
self.normalisation[sample] = self.initial_normalisation[sample]
self.background_subtraction(histograms)
# next, let's round all numbers (they are event numbers after all
for sample, values in self.normalisation.items():
new_values = [(round(v, 1), round(e, 1)) for v, e in values]
self.normalisation[sample] = new_values
self.have_normalisation = True
def calculate_normalisation(self):
'''
1. get file names
2. get histograms from files
3. ???
4. calculate normalisation
'''
if self.have_normalisation:
return
histograms = self.measurement.histograms
for sample, hist in histograms.items():
# TODO: this should be a list of bin-contents
hist = fix_overflow(hist)
histograms[sample] = hist
self.initial_normalisation[sample] = hist_to_value_error_tuplelist(hist)
self.normalisation[sample] = self.initial_normalisation[sample]
self.background_subtraction(histograms)
# next, let's round all numbers (they are event numbers after all
for sample, values in self.normalisation.items():
new_values = [(round(v, 1), round(e, 1)) for v, e in values]
self.normalisation[sample] = new_values
self.have_normalisation = True
def test_overflow_2D( self ):
before_fix = check_overflow_in_2DHist(self.h2)
has_overflow_in_x = before_fix['has_overflow_in_x']
has_overflow_in_y = before_fix['has_overflow_in_y']
self.assertGreater(has_overflow_in_x, 0, '2D hist: No overflow in x present, wrong setup.')
self.assertGreater(has_overflow_in_y, 0, '2D hist: No overflow in y present, wrong setup.')
h2 = fix_overflow( self.h2 )
after_fix = check_overflow_in_2DHist(h2)
has_overflow_in_x = after_fix['has_overflow_in_x']
has_overflow_in_y = after_fix['has_overflow_in_y']
# check if overflow has been reset
self.assertEqual( has_overflow_in_x, 0, '2D hist: Overflow in x is not 0.' )
self.assertEqual( has_overflow_in_y, 0, '2D hist: Overflow in y is not 0.' )
# now check if new last bin content is equal to the old one plus overflow
overflow_x_before = before_fix['overflow_x']
overflow_y_before = before_fix['overflow_y']
last_bin_content_x_before = before_fix['last_bin_content_x']
last_bin_content_y_before = before_fix['last_bin_content_y']
last_bin_content_x_after = after_fix['last_bin_content_x']
last_bin_content_y_after = after_fix['last_bin_content_y']
check_last_bin_content_x = [overflow + last_bin_content for overflow,last_bin_content in zip(overflow_x_before, last_bin_content_x_before)]
check_last_bin_content_y = [overflow + last_bin_content for overflow,last_bin_content in zip(overflow_y_before, last_bin_content_y_before)]
# remember, the last item in each list is actually the overflow, which should be 0 and the above calculation is not correct.
self.assertTrue(check_equal_lists(check_last_bin_content_x[:-2], last_bin_content_x_after[:-2]), '2D hist: last bins in x are not correct.')
self.assertTrue(check_equal_lists(check_last_bin_content_y[:-2], last_bin_content_y_after[:-2]), '2D hist: last bins in y are not correct.')
def setUp(self):
# create histograms
# self.h1 = Hist2D( 15, 0, 15, 15, 0, 15 )
# x_1 = [1, 3, 7, 7, 8, 1, 12, 7, 8, 6]
# y_1 = [1, 7, 3, 7, 8, 12, 7, 12, 13, 11]
self.h1 = Hist2D(60, 40, 100, 60, 40, 100)
n_1 = 100000
x_1 = 60 + 10 * np.random.randn(n_1)
y_1 = x_1 + np.random.randn(n_1)
z_1 = np.vstack((x_1, y_1)).T
self.h1.fill_array(z_1)
self.h1 = fix_overflow(self.h1)
self.histogram_information = [{"hist": self.h1, "CoM": 7, "channel": "test_1"}]
self.histograms = [info["hist"] for info in self.histogram_information]
# requirements for new binning
self.p_min, self.s_min, self.n_min = 0.5, 0.5, 1000
self.bin_edges = []
self.purities_GetBinContent = []
self.stabilities_GetBinContent = []
self.n_events_GetBinContent = []
self.purities_Integral = []
self.stabilities_Integral = []
self.n_events_Integral = []
first_hist = self.histograms[0]
n_bins = first_hist.GetNbinsX()
current_bin_start = 0
current_bin_end = 0
while current_bin_end < n_bins:
current_bin_end, p, s, n_gen_and_reco = pick_bins.get_next_end(
self.histograms, current_bin_start, current_bin_end, self.p_min, self.s_min, self.n_min, 0
)
if not self.bin_edges:
# if empty
self.bin_edges.append(first_hist.GetXaxis().GetBinLowEdge(current_bin_start + 1))
self.bin_edges.append(
first_hist.GetXaxis().GetBinLowEdge(current_bin_end)
+ first_hist.GetXaxis().GetBinWidth(current_bin_end)
)
self.purities_Integral.append(p)
self.stabilities_Integral.append(s)
self.n_events_Integral.append(n_gen_and_reco)
current_bin_start = current_bin_end
self.h1_rebinned = rebin_2d(self.h1, self.bin_edges, self.bin_edges)
self.purities_GetBinContent = calculate_purities(self.h1_rebinned)
self.stabilities_GetBinContent = calculate_stabilities(self.h1_rebinned)
self.n_events_GetBinContent = [int(self.h1_rebinned.GetBinContent(i, i)) for i in range(1, len(self.bin_edges))]
def __return_histogram(self, d_hist_info, ignoreUnderflow=True, useQCDControl=False, useQCDSystematicControl=False):
'''
Takes basic histogram info and returns histo.
Maybe this can move to ROOT_utilities?
'''
from rootpy.io.file import File
from rootpy.plotting import Hist
from dps.utils.hist_utilities import fix_overflow
f = d_hist_info['input_file']
tree = d_hist_info['tree']
qcd_tree = d_hist_info["qcd_control_region"]
qcd_tree_for_normalisation = d_hist_info["qcd_normalisation_region"]
var = d_hist_info['branch']
bins = d_hist_info['bin_edges']
lumi_scale = d_hist_info['lumi_scale']
scale = d_hist_info['scale']
weights = d_hist_info['weight_branches']
selection = d_hist_info['selection']
if useQCDControl:
# replace SR tree with CR tree
if useQCDSystematicControl:
tree = qcd_tree_for_normalisation
else:
tree = qcd_tree
# Remove the Lepton reweighting for the datadriven qcd (SF not derived for unisolated leptons)
for weight in weights:
if 'Electron' in weight: weights.remove(weight)
elif 'Muon' in weight: weights.remove(weight)
weights = "*".join(weights)
# Selection will return a weight 0 or 1 depending on whether event passes selection
weights_and_selection = '( {0} ) * ( {1} )'.format(weights, selection)
scale *= lumi_scale
root_file = File( f )
root_tree = root_file.Get( tree )
root_histogram = Hist( bins )
# Draw histogram of var for selection into root_histogram
root_tree.Draw(var, selection = weights_and_selection, hist = root_histogram)
root_histogram.Scale(scale)
# When a tree is filled with a dummy variable, it will end up in the underflow, so ignore it
if ignoreUnderflow:
root_histogram.SetBinContent(0, 0)
root_histogram.SetBinError(0,0)
# Fix overflow (Moves entries from overflow bin into last bin i.e. last bin not |..| but |--> )
root_histogram = fix_overflow(root_histogram)
root_file.Close()
return root_histogram
def test_overflow_1D( self ):
last_bin = self.h1.nbins()
overflow_bin = last_bin + 1
overflow = self.h1.GetBinContent( overflow_bin )
last_bin_content = self.h1.GetBinContent( last_bin )
self.assertGreater( overflow, 0, '1D hist: No overflow present, wrong setup.' )
h1 = fix_overflow( self.h1 )
self.assertEqual( h1.GetBinContent( overflow_bin ), 0., '1D hist: Overflow bin is not 0.' )
self.assertEqual( h1.GetBinContent( last_bin ), last_bin_content + overflow, '1D hist: last bin is not correct.' )
def test_overflow_1D(self):
last_bin = self.h1.nbins()
overflow_bin = last_bin + 1
overflow = self.h1.GetBinContent(overflow_bin)
last_bin_content = self.h1.GetBinContent(last_bin)
self.assertGreater(overflow, 0,
'1D hist: No overflow present, wrong setup.')
h1 = fix_overflow(self.h1)
self.assertEqual(h1.GetBinContent(overflow_bin), 0.,
'1D hist: Overflow bin is not 0.')
self.assertEqual(h1.GetBinContent(last_bin),
last_bin_content + overflow,
'1D hist: last bin is not correct.')
def test_overflow_2D(self):
before_fix = check_overflow_in_2DHist(self.h2)
has_overflow_in_x = before_fix['has_overflow_in_x']
has_overflow_in_y = before_fix['has_overflow_in_y']
self.assertGreater(has_overflow_in_x, 0,
'2D hist: No overflow in x present, wrong setup.')
self.assertGreater(has_overflow_in_y, 0,
'2D hist: No overflow in y present, wrong setup.')
h2 = fix_overflow(self.h2)
after_fix = check_overflow_in_2DHist(h2)
has_overflow_in_x = after_fix['has_overflow_in_x']
has_overflow_in_y = after_fix['has_overflow_in_y']
# check if overflow has been reset
self.assertEqual(has_overflow_in_x, 0,
'2D hist: Overflow in x is not 0.')
self.assertEqual(has_overflow_in_y, 0,
'2D hist: Overflow in y is not 0.')
# now check if new last bin content is equal to the old one plus overflow
overflow_x_before = before_fix['overflow_x']
overflow_y_before = before_fix['overflow_y']
last_bin_content_x_before = before_fix['last_bin_content_x']
last_bin_content_y_before = before_fix['last_bin_content_y']
last_bin_content_x_after = after_fix['last_bin_content_x']
last_bin_content_y_after = after_fix['last_bin_content_y']
check_last_bin_content_x = [
overflow + last_bin_content for overflow, last_bin_content in zip(
overflow_x_before, last_bin_content_x_before)
]
check_last_bin_content_y = [
overflow + last_bin_content for overflow, last_bin_content in zip(
overflow_y_before, last_bin_content_y_before)
]
# remember, the last item in each list is actually the overflow, which should be 0 and the above calculation is not correct.
self.assertTrue(
check_equal_lists(check_last_bin_content_x[:-2],
last_bin_content_x_after[:-2]),
'2D hist: last bins in x are not correct.')
self.assertTrue(
check_equal_lists(check_last_bin_content_y[:-2],
last_bin_content_y_after[:-2]),
'2D hist: last bins in y are not correct.')
def setUp(self):
# create histograms
# self.h1 = Hist2D( 15, 0, 15, 15, 0, 15 )
# x_1 = [1, 3, 7, 7, 8, 1, 12, 7, 8, 6]
# y_1 = [1, 7, 3, 7, 8, 12, 7, 12, 13, 11]
self.h1 = Hist2D(60, 40, 100, 60, 40, 100)
n_1 = 100000
x_1 = 60 + 10 * np.random.randn(n_1)
y_1 = x_1 + np.random.randn(n_1)
z_1 = np.vstack((x_1, y_1)).T
self.h1.fill_array(z_1)
self.h1 = fix_overflow(self.h1)
self.histogram_information = [
{
'hist': self.h1,
'CoM': 7,
'channel': 'test_1'
},
]
self.histograms = [info['hist'] for info in self.histogram_information]
# requirements for new binning
self.p_min, self.s_min, self.n_min = 0.5, 0.5, 1000
self.bin_edges = []
self.purities_GetBinContent = []
self.stabilities_GetBinContent = []
self.n_events_GetBinContent = []
self.purities_Integral = []
self.stabilities_Integral = []
self.n_events_Integral = []
first_hist = self.histograms[0]
n_bins = first_hist.GetNbinsX()
current_bin_start = 0
current_bin_end = 0
while current_bin_end < n_bins:
current_bin_end, p, s, n_gen_and_reco = pick_bins.get_next_end(
self.histograms, current_bin_start, current_bin_end,
self.p_min, self.s_min, self.n_min, 0)
if not self.bin_edges:
# if empty
self.bin_edges.append(
first_hist.GetXaxis().GetBinLowEdge(current_bin_start + 1))
self.bin_edges.append(
first_hist.GetXaxis().GetBinLowEdge(current_bin_end) +
first_hist.GetXaxis().GetBinWidth(current_bin_end))
self.purities_Integral.append(p)
self.stabilities_Integral.append(s)
self.n_events_Integral.append(n_gen_and_reco)
current_bin_start = current_bin_end
self.h1_rebinned = rebin_2d(self.h1, self.bin_edges, self.bin_edges)
self.purities_GetBinContent = calculate_purities(self.h1_rebinned)
self.stabilities_GetBinContent = calculate_stabilities(
self.h1_rebinned)
self.n_events_GetBinContent = [
int(self.h1_rebinned.GetBinContent(i, i))
for i in range(1, len(self.bin_edges))
]
def __return_histogram(self,
d_hist_info,
ignoreUnderflow=True,
useQCDControl=False,
useQCDSystematicControl=False):
'''
Takes basic histogram info and returns histo.
Maybe this can move to ROOT_utilities?
'''
from rootpy.io.file import File
from rootpy.plotting import Hist
from dps.utils.hist_utilities import fix_overflow
f = d_hist_info['input_file']
tree = d_hist_info['tree']
qcd_tree = d_hist_info["qcd_control_region"]
qcd_tree_for_normalisation = d_hist_info["qcd_normalisation_region"]
var = d_hist_info['branch']
bins = d_hist_info['bin_edges']
lumi_scale = d_hist_info['lumi_scale']
scale = d_hist_info['scale']
weights = d_hist_info['weight_branches']
selection = d_hist_info['selection']
if useQCDControl:
# replace SR tree with CR tree
if useQCDSystematicControl:
tree = qcd_tree_for_normalisation
else:
tree = qcd_tree
# Remove the Lepton reweighting for the datadriven qcd (SF not derived for unisolated leptons)
for weight in weights:
if 'Electron' in weight: weights.remove(weight)
elif 'Muon' in weight: weights.remove(weight)
weights = "*".join(weights)
# Selection will return a weight 0 or 1 depending on whether event passes selection
weights_and_selection = '( {0} ) * ( {1} )'.format(weights, selection)
scale *= lumi_scale
root_file = File(f)
root_tree = root_file.Get(tree)
root_histogram = Hist(bins)
# Draw histogram of var for selection into root_histogram
root_tree.Draw(var,
selection=weights_and_selection,
hist=root_histogram)
root_histogram.Scale(scale)
# When a tree is filled with a dummy variable, it will end up in the underflow, so ignore it
if ignoreUnderflow:
root_histogram.SetBinContent(0, 0)
root_histogram.SetBinError(0, 0)
# Fix overflow (Moves entries from overflow bin into last bin i.e. last bin not |..| but |--> )
root_histogram = fix_overflow(root_histogram)
root_file.Close()
return root_histogram
