8.5. FAQ#

8.5.1. PID corrections and systematic errors#

Efficiency correction factors and systematic errors for Belle PID were obtained using corresponding MC and data samples. Those studies were done for different PID selections, and possible selections differ for different hypotheses.

There are three groups of lookup tables in Belle:

Lepton ID:
- Possible cuts for electron: [0.01, 0.10, 0.50, 0.60, 0.80, 0.90]
- Possible cuts for muons: [0.10, 0.80, 0.90, 0.95, 0.97]
Kaon/\(\pi\) ID:
- PID cut can be 0.1 - 0.9
Proton ID:
- Possible cuts for proton: [0.6, 0.7, 0.8, 0.9]

Details of these tables and how they were studies can be found at Belle PID Joint Homepage.

The weights for LID and KID are available in basf2 and can be retrieved via payloads in Belle_PID; naming schemes of each PID payloads are slightly different.

Lepton ID

Naming scheme for LID is BelleLID<e|mu>_<LID cut>. Using eIDBelle>0.9 as an example:

import variables as va
import basf2 as b2

b2.conditions.prepend_globaltag("BellePID")

lid_table = "BelleLIDe_0.90"
va.variables.addAlias('LID_ratio', f"extraInfo({lid_table}_ratio)")
va.variables.addAlias('LID_ratio_stat_err', f"extraInfo({lid_table}_ratio_stat_err)")
va.variables.addAlias('LID_ratio_syst_err', f"extraInfo({lid_table}_ratio_syst_err)")
va.variables.addAlias('LID_ratio_syst_err_from_2photon', f"extraInfo({lid_table}_ratio_syst_err_from_2photon)")
va.variables.addAlias('LID_ratio_syst_err_from_JPsi', f"extraInfo({lid_table}_ratio_syst_err_from_JPsi)")
va.variables.addAlias('LID_run_bin', f"extraInfo({lid_table}_run_bin)")
va.variables.addAlias('LID_kinematic_bin', f"extraInfo({lid_table}_kinematic_bin)")

reweighter = b2.register_module('ParticleWeighting')
reweighter.param('tableName', lid_table)
reweighter.param('particleList', 'e+:all')
my_path.add_module(reweighter)

Note

The lepton ID fake-rate tables (lepton ID efficiency for pions and kaons) are not prepared at Belle. It can be studied using the \(D^{*}\) sample (see ~nishida6/public/kid/00dstar_nt.txt on kekcc for details).

Kaon/\(\pi\) ID

Naming scheme for payloads is: BelleKID<K|Pi><Eff|Fake><Combined|Plus|Minus>_<PID cut:[1..9]> Here,

K or Pi refers to correction information obtained for kaons or pions, accordingly
Eff or Fake refers to efficiency or fake rate corrections
Combined, Plus or Minus refers to the charge of tracks used for the study
PID cut is an integer number from 1 to 9 that corresponds to “cut > 0.number”

To append PID correction information for pion efficiency selected as kaons with atcPIDBelle(3, 2)>0.6 cut:

kid_table = "BelleKIDPiFakePlus_6"

va.variables.addAlias('KID_eff_data', f"extraInfo({kid_table}_eff_data)")
va.variables.addAlias('KID_eff_data_stat_err', f"extraInfo({kid_table}_eff_data_stat_err)")
va.variables.addAlias('KID_eff_data_syst_err', f"extraInfo({kid_table}_eff_data_syst_err)")
va.variables.addAlias('KID_eff_mc', f"extraInfo({kid_table}_eff_mc)")
va.variables.addAlias('KID_eff_mc_stat_err', f"extraInfo({kid_table}_eff_mc_stat_err)")
va.variables.addAlias('KID_fit_flag', f"extraInfo({kid_table}_fit_flag)")
va.variables.addAlias('KID_kinematic_bin', f"extraInfo({kid_table}_kinematic_bin)")
va.variables.addAlias('KID_ratio', f"extraInfo({kid_table}_ratio)")
va.variables.addAlias('KID_ratio_stat_err', f"extraInfo({kid_table}_ratio_stat_err)")
va.variables.addAlias('KID_ratio_syst_err', f"extraInfo({kid_table}_ratio_syst_err)")
va.variables.addAlias('KID_run_bin', f"extraInfo({kid_table}_run_bin)")

kid_weights = ['KID_eff_data',
               'KID_eff_data_stat_err',
               'KID_eff_data_syst_err',
               'KID_eff_mc',
               'KID_eff_mc_stat_err',
               'KID_fit_flag',
               'KID_ratio',
               'KID_ratio_stat_err',
               'KID_ratio_syst_err',
               'KID_run_bin',
               'KID_kinematic_bin']

reweighter2 = b2.register_module('ParticleWeighting')
reweighter2.param('tableName', kid_table)
reweighter2.param('particleList', 'pi+:all')
my_path.add_module(reweighter2)

Note

fit_flag indicates if the fits failed; fit_flag = 0 is for successful fit, while any non-zero values of fit_flag are for failed fit. Please only use fit_flag==0 when evaluating PID contributions.

Proton ID

Naming scheme for proton ID payloads is: BellePID<Plus|Minus>_<PID cut>

Here,

Plus or Minus refers to the charge of tracks used for the study
PID cut can be 0.6, 0.7, 0.8 or 0.9

If you’d like to add the information for proton efficiency selected with atcPIDBelle(4, 2)>0.6 and atcPIDBelle(4, 3)>0.6.

pid_table = "BellePIDPlus_0.6"

va.variables.addAlias("PID_eff_data", f"extraInfo({pid_table}_eff_data)")
va.variables.addAlias("PID_eff_data_stat_err", f"extraInfo({pid_table}_eff_data_stat_err)")
va.variables.addAlias("PID_eff_mc", f"extraInfo({pid_table}_eff_mc)")
va.variables.addAlias("PID_eff_mc_stat_err", f"extraInfo({pid_table}_eff_mc_stat_err)")
va.variables.addAlias("PID_ratio", f"extraInfo({pid_table}_ratio)")
va.variables.addAlias("PID_ratio_stat_err", f"extraInfo({pid_table}_ratio_stat_err)")
va.variables.addAlias("PID_ratio_syst_err", f"extraInfo({pid_table}_ratio_syst_err)")
va.variables.addAlias("PID_kinematic_bin", f"extraInfo({pid_table}_kinematic_bin)")
va.variables.addAlias("PID_run_bin", f"extraInfo({pid_table}_run_bin)")

reweighter3 = b2.register_module('ParticleWeighting')
reweighter3.param('tableName', pid_table)
reweighter3.param('particleList', 'pi+:all')
my_path.add_module(reweighter3)

8.5.2. Full Event Interpretation#

To utilize FEI, the correct prefix of FEI payloads needs to be set:

import fei
basf2.conditions.append_globaltag("analysis_tools_light-2012-minos")
configuration = fei.config.FeiConfiguration(prefix='FEI_B2BII_light-2012-minos', cache=0)

feistate = fei.get_path(particles, configuration)

path.add_path(feistate.path)

For more details please see analysis/examples/FEI/B_converted_apply.py

8.5.3. Flavour Tagger#

To apply flavour tagger in a b2bii analysis, one will need to append the correct global tag. FlavorTagger will call the corresponding payloads in the module.

import flavorTagger as ft

# Flavour Tagger
weightfiles = 'B2nunubarBGx1'
basf2.conditions.append_globaltag("analysis_tools_light-2012-minos")
ft.flavorTagger(
    particleLists=['B+:sig'],
    weightFiles=weightfiles,
    prefix='',
    path=my_path)

8.5.4. BeamBackground MVA and FakePhoton MVA#

The weight files for the b2bii case are stored in the analysis_b2bii global tag. To apply beam background MVA and fake photon MVA in a b2bii analysis, the correct suffix, Belle, should be used for the weight file, and apply them over the premade gamma:mdst list:

import modularAnalysis as ma
basf2.conditions.append_globaltag(ma.getAnalysisGlobaltagB2BII())
# Apply fake photon MVA
ma.getFakePhotonProbability(
    particleList=['gamma:mdst'],
    weight="Belle",
    path=mypath)
# Apply beamBackground MVA
ma.getBeamBackgroundProbability(
    particleList=['gamma:mdst'],
    weight="Belle",
    path=mypath)

Note

Details for the training done in Belle MC can be found in the appendix in BELLE2-NOTE-PH-2023-036.

basf2 development documentation

FAQ

Contents

8.5. FAQ#

8.5.1. PID corrections and systematic errors#

8.5.2. Full Event Interpretation#

8.5.3. Flavour Tagger#

8.5.4. BeamBackground MVA and FakePhoton MVA#