Skip to content

bugfixes and feature additions #29

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
a-monsch wants to merge 21 commits into
base: main
Choose a base branch
from
Open

bugfixes and feature additions #29

Show file tree
Hide file tree
Changes from 10 commits
Commits
Show all changes
21 commits
Select commit Hold shift + click to select a range
ba9c770
fixing bin-by-bin subtraction for correction estimation + first draft…
a-monsch Mar 3, 2026
80418e5
fixing missing attribute
a-monsch Mar 4, 2026
d557628
enabling optionality of classic/model FF evaluators depending on
a-monsch Mar 4, 2026
bb0648f
adjusted correction pruning
a-monsch Mar 5, 2026
9f75725
testing other p_value estimates
a-monsch Mar 6, 2026
c103368
adjustmetns to optional DR_SR derivations
a-monsch Mar 9, 2026
1da198f
adding updated smhtt_ul 2018 config
a-monsch Mar 12, 2026
2f72f7e
minor adjsutmetns for sctatistical test procedure for correction sele…
a-monsch Mar 13, 2026
45bc1a5
minor adjustments, fraction uncertainty fix
a-monsch Apr 6, 2026
0a09aef
Merge branch 'main' into sm_devs
nshadskiy Apr 16, 2026
23cb78c
removed dy selection
sofiagiappichini Apr 17, 2026
b84897b
revert change regarding DR_SR fake factor related config parameters
nshadskiy Apr 17, 2026
2122e69
update config
nshadskiy Apr 17, 2026
b9c64c7
fix some bugs
nshadskiy Apr 20, 2026
4f2437b
update documentation
nshadskiy Apr 20, 2026
73049fe
config updates
nshadskiy Apr 20, 2026
513519c
update documentation
nshadskiy Apr 21, 2026
9c8a8dc
update conda environment
nshadskiy Apr 21, 2026
abe22c8
fix 2D FF write out to correctionlib
nshadskiy Apr 27, 2026
f4d42b9
fix cornor case of fit_function
nshadskiy Apr 27, 2026
80223a5
fix index
nshadskiy Apr 27, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions .gitignore
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,7 @@ configs.*/
*.pdf
*.png
*.root
*.feather

# C extensions
*.so
Expand Down
269 changes: 106 additions & 163 deletions FF_calculation/FF_QCD.py
View file
Open in desktop

Large diffs are not rendered by default.

333 changes: 137 additions & 196 deletions FF_calculation/FF_Wjets.py
View file
Open in desktop

Large diffs are not rendered by default.

145 changes: 38 additions & 107 deletions FF_calculation/FF_ttbar.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -54,7 +54,6 @@ def calculation_ttbar_FFs(

log = logging.getLogger(logger)

# init histogram dict for FF measurement from MC
SR_hists = dict()
AR_hists = dict()

Expand All @@ -70,10 +69,7 @@ def calculation_ttbar_FFs(

rdf = ROOT.RDataFrame(config["tree"], sample_path)

# event filter for ttbar signal region
log.info(
f"Filtering events for the signal region. Target process: {process}"
)
log.info(f"Filtering events for the signal region. Target process: {process}")
region_conf = copy.deepcopy(process_conf["SR_cuts"])
rdf_SR = ff_func.apply_region_filters(
rdf=rdf,
Expand All @@ -84,10 +80,7 @@ def calculation_ttbar_FFs(
logger=logger,
)

# event filter for ttbar application region
log.info(
f"Filtering events for the application region. Target process: {process}"
)
log.info(f"Filtering events for the application region. Target process: {process}")
region_conf = copy.deepcopy(process_conf["AR_cuts"])
rdf_AR = ff_func.apply_region_filters(
rdf=rdf,
Expand All @@ -98,34 +91,21 @@ def calculation_ttbar_FFs(
logger=logger,
)

# get binning of the dependent variable
xbinning = array.array("d", splitting.var_bins)
nbinsx = len(splitting.var_bins) - 1

# making the histograms
h = RuntimeVariables.RDataFrameWrapper(rdf_SR).Histo1D(
(
process_conf["var_dependence"],
f"{sample}",
nbinsx,
xbinning,
),
SR_hists[sample] = RuntimeVariables.RDataFrameWrapper(rdf_SR).Histo1D(
(process_conf["var_dependence"], f"{sample}", nbinsx, xbinning),
process_conf["var_dependence"],
"weight",
)
SR_hists[sample] = h.GetValue()

h = RuntimeVariables.RDataFrameWrapper(rdf_AR).Histo1D(
(
process_conf["var_dependence"],
f"{sample}",
nbinsx,
xbinning,
),
).GetValue()

AR_hists[sample] = RuntimeVariables.RDataFrameWrapper(rdf_AR).Histo1D(
(process_conf["var_dependence"], f"{sample}", nbinsx, xbinning),
process_conf["var_dependence"],
"weight",
)
AR_hists[sample] = h.GetValue()
).GetValue()

# Start of the FF calculation
FF_hist = ff_func.calculate_ttbar_FF(
Expand All @@ -142,14 +122,15 @@ def calculation_ttbar_FFs(
logger=logger,
fit_option=splitting.fit_option,
limit_kwargs=splitting.limit_kwargs(hist=FF_hist),
stat_sigma=config["stat_sigma"] if "stat_sigma" in config else 1.0,
)
elif isinstance(splitting.fit_option, str):
nominal_draw_obj, results = ff_func.smooth_function(
hist=FF_hist.Clone(),
bin_edges=splitting.var_bins,
correction_option=splitting.fit_option,
bandwidth=splitting.bandwidth,
for_FF=True,
stat_sigma=config["stat_sigma"] if "stat_sigma" in config else 1.0,
)
unc_draw_obj = results["default"]
used_fit = splitting.fit_option
Expand All @@ -172,20 +153,10 @@ def calculation_ttbar_FFs(
save_data=True,
)

# doing some control plots
# doing control plots
for _hist, _region, _data, _samples in [
(
SRlike_hists,
"SR_like",
"data",
ff_func.controlplot_samples(config["use_embedding"]),
),
(
ARlike_hists,
"AR_like",
"data",
ff_func.controlplot_samples(config["use_embedding"]),
),
(SRlike_hists, "SR_like", "data", ff_func.controlplot_samples(sample_paths)),
(ARlike_hists, "AR_like", "data", ff_func.controlplot_samples(sample_paths)),
(SRlike_hists, "SR_like", "data_subtracted", ["ttbar_J"]),
(ARlike_hists, "AR_like", "data_subtracted", ["ttbar_J"]),
]:
Expand Down Expand Up @@ -215,11 +186,7 @@ def calculation_ttbar_FFs(
return results


@logging_helper.LogDecorator().grouped_logs(
extractor=lambda *args, **kwargs: (
kwargs["logger"] if "config" in kwargs else args[4]
)
)
@logging_helper.LogDecorator().grouped_logs(extractor=lambda *args, **kwargs: (kwargs["logger"] if "config" in kwargs else args[4]))
def calculation_FF_data_scaling_factor(
config: Dict[str, Union[str, Dict, List]],
process_conf: Dict[str, Union[str, Dict, List]],
Expand All @@ -244,10 +211,8 @@ def calculation_FF_data_scaling_factor(
"""
log = logging.getLogger(logger)

# init histogram dict for FF data correction
SRlike_hists = dict()
ARlike_hists = dict()
# init histogram dict for QCD SS/OS estimation
SRlike_hists_qcd = dict()
ARlike_hists_qcd = dict()

Expand All @@ -259,10 +224,7 @@ def calculation_FF_data_scaling_factor(

rdf = ROOT.RDataFrame(config["tree"], sample_path)

# event filter for ttbar signal-like region
log.info(
f"Filtering events for the signal-like region. Target process: {process}"
)
log.info(f"Filtering events for the signal-like region. Target process: {process}")
region_conf = copy.deepcopy(process_conf["SRlike_cuts"])
rdf_SRlike = ff_func.apply_region_filters(
rdf=rdf,
Expand All @@ -273,8 +235,7 @@ def calculation_FF_data_scaling_factor(
logger=logger,
)

# QCD estimation from same sign in signal-like region
if "tau_pair_sign" in region_conf:
if "tau_pair_sign" in region_conf: # QCD estimation from same sign in signal-like region
region_conf["tau_pair_sign"] = "(q_1*q_2) > 0" # same sign
else:
raise ValueError(
Expand All @@ -293,10 +254,7 @@ def calculation_FF_data_scaling_factor(
logger=logger,
)

# event filter for ttbar application-like region
log.info(
f"Filtering events for the application-like region. Target process: {process}"
)
log.info(f"Filtering events for the application-like region. Target process: {process}")
region_conf = copy.deepcopy(process_conf["ARlike_cuts"])
rdf_ARlike = ff_func.apply_region_filters(
rdf=rdf,
Expand All @@ -307,8 +265,7 @@ def calculation_FF_data_scaling_factor(
logger=logger,
)

# QCD estimation from same sign in application-like region
if "tau_pair_sign" in region_conf:
if "tau_pair_sign" in region_conf: # QCD estimation from same sign in application-like region
region_conf["tau_pair_sign"] = "(q_1*q_2) > 0" # same sign
else:
raise ValueError(
Expand All @@ -328,26 +285,22 @@ def calculation_FF_data_scaling_factor(
)

# make yield histograms for FF data correction
h = RuntimeVariables.RDataFrameWrapper(rdf_SRlike).Histo1D(
SRlike_hists[sample] = RuntimeVariables.RDataFrameWrapper(rdf_SRlike).Histo1D(
("#phi(#slash{E}_{T})", f"{sample}", 1, -3.5, 3.5), "metphi", "weight"
)
SRlike_hists[sample] = h.GetValue()
).GetValue()

h = RuntimeVariables.RDataFrameWrapper(rdf_ARlike).Histo1D(
ARlike_hists[sample] = RuntimeVariables.RDataFrameWrapper(rdf_ARlike).Histo1D(
("#phi(#slash{E}_{T})", f"{sample}", 1, -3.5, 3.5), "metphi", "weight"
)
ARlike_hists[sample] = h.GetValue()
).GetValue()

# make yield histograms for QCD estimation
h_qcd = RuntimeVariables.RDataFrameWrapper(rdf_SRlike_qcd).Histo1D(
SRlike_hists_qcd[sample] = RuntimeVariables.RDataFrameWrapper(rdf_SRlike_qcd).Histo1D(
("#phi(#slash{E}_{T})", f"{sample}", 1, -3.5, 3.5), "metphi", "weight"
)
SRlike_hists_qcd[sample] = h_qcd.GetValue()
).GetValue()

h_qcd = RuntimeVariables.RDataFrameWrapper(rdf_ARlike_qcd).Histo1D(
ARlike_hists_qcd[sample] = RuntimeVariables.RDataFrameWrapper(rdf_ARlike_qcd).Histo1D(
("#phi(#slash{E}_{T})", f"{sample}", 1, -3.5, 3.5), "metphi", "weight"
)
ARlike_hists_qcd[sample] = h_qcd.GetValue()
).GetValue()

# calculate QCD estimation
SRlike_hists["QCD"] = ff_func.QCD_SS_estimate(hists=SRlike_hists_qcd)
Expand Down Expand Up @@ -412,7 +365,6 @@ def non_closure_correction(

log = logging.getLogger(logger)

# init histogram dict for FF measurement
SR_hists = dict()
AR_hists = dict()

Expand All @@ -432,10 +384,7 @@ def non_closure_correction(

rdf = ROOT.RDataFrame(config["tree"], sample_path)

# event filter for ttbar signal region
log.info(
f"Filtering events for the signal region. Target process: {process}"
)
log.info(f"Filtering events for the signal region. Target process: {process}")
region_conf = copy.deepcopy(config["target_processes"][process]["SR_cuts"])
rdf_SR = ff_func.apply_region_filters(
rdf=rdf,
Expand All @@ -446,10 +395,7 @@ def non_closure_correction(
logger=logger,
)

# event filter for ttbar application region
log.info(
f"Filtering events for the application region. Target process: {process}"
)
log.info(f"Filtering events for the application region. Target process: {process}")
region_conf = copy.deepcopy(config["target_processes"][process]["AR_cuts"])
rdf_AR = ff_func.apply_region_filters(
rdf=rdf,
Expand All @@ -465,43 +411,27 @@ def non_closure_correction(
# additionally evaluate the previous corrections
corr_str = ""
for corr_evaluator in corr_evaluators:
rdf_AR = corr_evaluator.evaluate_correction(
rdf=rdf_AR,
)
rdf_AR = corr_evaluator.evaluate_correction(rdf=rdf_AR)
corr_str += f" * {corr_evaluator.corr_str}"

rdf_AR = rdf_AR.Define(
"weight_ff", f"weight * {process}_fake_factor{corr_str}"
)

# get binning of the dependent variable
xbinning = array.array("d", splitting.var_bins)
nbinsx = len(splitting.var_bins) - 1

# making the histograms
h = RuntimeVariables.RDataFrameWrapper(rdf_SR).Histo1D(
(
correction_conf["var_dependence"],
f"{sample}",
nbinsx,
xbinning,
),
SR_hists[sample] = RuntimeVariables.RDataFrameWrapper(rdf_SR).Histo1D(
(correction_conf["var_dependence"], f"{sample}", nbinsx, xbinning),
correction_conf["var_dependence"],
"weight",
)
SR_hists[sample] = h.GetValue()

h = RuntimeVariables.RDataFrameWrapper(rdf_AR).Histo1D(
(
correction_conf["var_dependence"],
f"{sample}",
nbinsx,
xbinning,
),
).GetValue()

AR_hists["ttbar_ff"] = RuntimeVariables.RDataFrameWrapper(rdf_AR).Histo1D(
(correction_conf["var_dependence"], f"{sample}", nbinsx, xbinning),
correction_conf["var_dependence"],
"weight_ff",
)
AR_hists["ttbar_ff"] = h.GetValue()
).GetValue()

correction_hist = ff_func.calculate_non_closure_correction_ttbar_fromMC(
SR=SR_hists, AR=AR_hists
Expand All @@ -512,6 +442,7 @@ def non_closure_correction(
bin_edges=splitting.var_bins,
correction_option=splitting.correction_option,
bandwidth=splitting.bandwidth,
stat_sigma=config["stat_sigma"] if "stat_sigma" in config else 1.0,
)

plotting.plot_correction(
Expand Down
Loading