ExpertTrackingValidationModule Class Reference

Inheritance diagram for ExpertTrackingValidationModule:

Collaboration diagram for ExpertTrackingValidationModule:

Public Member Functions
def	__init__ (self, name, contact, fit=False, pulls=False, resolution=False, output_file_name=None, track_filter_object=AlwaysPassFilter(), plot_name_postfix='', plot_title_postfix='', exclude_profile_mc_parameter='', exclude_profile_pr_parameter='', use_expert_folder=True, trackCandidatesColumnName='RecoTracks', mcTrackCandidatesColumnName='MCRecoTracks', cdcHitsColumnName='CDCHits', write_tables=False)
def	initialize (self)
def	event (self)
def	examine_hits_in_event (self)
def	terminate (self)
def	examine_pr_tracks (self)
def	examine_mc_tracks (self)
def	profiles_by_mc_parameters (self, xs, quantity_name, unit=None, parameter_names=['d_0', 'p_t', 'tan_lambda', 'multiplicity', 'phi', 'ndf',], make_hist=True, weights=None, is_asymmetry=False)
def	profiles_by_pr_parameters (self, xs, quantity_name, unit=None, parameter_names=['Seed_p_t', 'Seed tan #lambda', 'Seed #phi'], make_hist=True)
def	profiles_by_parameters_base (self, xs, quantity_name, parameter_names, profile_parameters, unit, make_hist, weights=None, is_asymmetry=False)

Public Attributes
	cdcHitsColumnname
	cached name of the CDCHits StoreArray
	write_tables
	cached value of the flag to write the validation figures of merit
	number_of_total_hits
	number of all hits
	number_of_mc_hits
	number of hits on MC track
	number_of_pr_hits
	number of hits on pattern reconstructed tracks
	is_hit_found
	list of flags for [not-]found hits
	is_hit_matched
	list of flags for [not-]matched hits
	mc_missing
	list of flags where MCRecoTrack is [not] missing MCTrackCand
	ratio_hits_in_mc_tracks_and_not_in_pr_tracks
	list of fraction of number of hits in MC track but not in PR track
	ratio_hits_in_mc_tracks_and_in_pr_tracks
	list of fraction of number of hits in MC track and in PR track
	ratio_hits_in_missing_mc_tracks_and_in_pr_tracks
	list of fraction of number of hits in missing MC track and in PR track
	ratio_hits_in_mc_tracks_and_in_fake_pr_tracks
	list of fraction of number of hits in MC track and in fake PR track
	ratio_hits_in_mc_tracks_and_in_good_pr_tracks
	list of fraction of number of hits in MC track and in good PR track
	mc_is_primary
	list of flags indicating that the MC track is [not] a primary MCParticle
	mc_number_of_hits
	list of the number of MCTrackCandHits on the MC track
	number_of_connected_tracks
	This is the number of mcTrackCands sharing a hit with the track cand.
	number_of_wrong_hits
	This number gives information about the "badness" of the fake.
	pr_number_of_hits
	list of the number of pattern-reconstructed hits
	pr_number_of_matched_hits
	list of the number of pattern-reconstructed hits matched to MC track
	validation_name
	cached value of the tracking-validation name
	contact
	cached value of the contact person name
	fit
	cached value of the track fit
	pulls
	cached values of the track-fit pulls
	resolution
	cached value of the resolution
	output_file_name
	cached value of the output ROOT TFile
	track_filter_object
	cached value of the track-filter object
	plot_name_postfix
	cached value of the suffix appended to the plot names
	plot_title_postfix
	cached value of the suffix appended to the plot titles
	exclude_profile_pr_parameter
	cached list of perigee parameters excluded from PR side plots
	exclude_profile_mc_parameter
	cached list of perigee parameters excluded from MC side plots
	use_expert_folder
	cached flag to use the "expert" folder for the pull and residual plots
	trackCandidatesColumnName
	cached name of the RecoTracks StoreArray
	mcTrackCandidatesColumnName
	cached name of the MCRecoTracks StoreArray
	non_expert_parameters
	list of parameters that determines which plots (all with corresponding x-axis) are marked as shifter plots
	resolution_pt_binning
	default binning used for resolution plots over pt
	referenceFileName
	If this variable is set the code will open the file with same name as the file created here and will read the binning from the TH1/TProfile with same name as the one created here. More...
	trackMatchLookUp
	Track-match object that examines relation information from MCMatcherTracksModule.
	pr_clones_and_matches
	Use deques in favour of lists to prevent repeated memory allocation of cost O(n) More...
	pr_matches
	list of PR-track matches
	pr_fakes
	list of PR-track fakes
	pr_seed_pt
	list of PR-track seed pt values
	pr_seed_tan_lambdas
	list of PR-track seed tan(lambda) values
	pr_seed_phi
	list of PR-track seed phi values
	pr_omega_truths
	list of PR-track seed omega-truth values
	pr_omega_estimates
	list of PR-track seed omega-estimate values
	pr_omega_variances
	list of PR-track seed omega-variance values
	pr_tan_lambda_truths
	list of PR-track seed tan(lambda)-truth values
	pr_tan_lambda_estimates
	list of PR-track seed tan(lambda)-estimate values
	pr_tan_lambda_variances
	list of PR-track seed tan(lambda)-variance values
	pr_d0_truths
	list of PR-track seed d0-truth values
	pr_d0_estimates
	list of PR-track seed d0-estimate values
	pr_d0_variances
	list of PR-track seed d0-variance values
	pr_z0_truths
	list of PR-track seed z0-truth values
	pr_z0_estimates
	list of PR-track seed z0-estimate values
	pr_pt_truths
	list of PR-track seed pt-truth values
	pr_pt_estimates
	list of PR-track seed pt-estimate values
	pr_bining_pt
	list of PR-track binning values
	mc_matches
	list of MC-track matches
	mc_charge_matches
	list of MC-track matches, including matched charge
	mc_charge_asymmetry
	list of MC-track matches charge asymmetry
	mc_charge_asymmetry_weights
	list of MC-track matches charge asymmetry weights
	mc_primaries
	list of MC-track primaries
	mc_d0s
	list of MC-track d0 values
	mc_tan_lambdas
	list of MC-track tan(lambda) values
	mc_phi
	direction of the track in phi
	mc_pts
	list of MC-track pt values
	mc_hit_efficiencies
	list of MC-track hit efficiencies
	mc_multiplicities
	list of MC-track multiplicities
	mc_ndf
	list of MC-track number of degrees of freedom

Detailed Description

Module to collect more matching information about the found particles and to generate validation
plots and figures of merit on the performance of track finding. This module gives information on the
number of hits etc. 

Definition at line 40 of file hit_module.py.

Constructor & Destructor Documentation

__init__()

def __init__	(		self,
			name,
			contact,
			fit = False,
			pulls = False,
			resolution = False,
			output_file_name = None,
			track_filter_object = AlwaysPassFilter(),
			plot_name_postfix = '',
			plot_title_postfix = '',
			exclude_profile_mc_parameter = '',
			exclude_profile_pr_parameter = '',
			use_expert_folder = True,
			trackCandidatesColumnName = 'RecoTracks',
			mcTrackCandidatesColumnName = 'MCRecoTracks',
			cdcHitsColumnName = 'CDCHits',
			write_tables = False
	)

Constructor

Definition at line 46 of file hit_module.py.

            write_tables=False):
        """Constructor"""
 
        TrackingValidationModule.__init__(
            self,
            name,
            contact,
            fit,
            pulls,
            resolution,
            output_file_name,
            track_filter_object,
            plot_name_postfix,
            plot_title_postfix,
            exclude_profile_mc_parameter,
            exclude_profile_pr_parameter,
            use_expert_folder,
            trackCandidatesColumnName,
            mcTrackCandidatesColumnName)
 
        
        self.cdcHitsColumnname = cdcHitsColumnName
        
        self.write_tables = write_tables
 

Member Function Documentation

event()

def event

(

self

)

Event method

Reimplemented from TrackingValidationModule.

Definition at line 137 of file hit_module.py.

examine_hits_in_event()

def examine_hits_in_event

(

self

)

Classify all of the hits in the event according to the parent track(s)

Definition at line 143 of file hit_module.py.

examine_mc_tracks()

def examine_mc_tracks (   self )

inherited

Looks at the individual Monte Carlo tracks and store information about them

Definition at line 382 of file module.py.

    def examine_mc_tracks(self):
        """Looks at the individual Monte Carlo tracks and store information about them"""
 
        trackMatchLookUp = self.trackMatchLookUp
 
        # Analyse from the Monte Carlo reference side
        mcTrackCands = Belle2.PyStoreArray(self.mcTrackCandidatesColumnName)
        mcParticles = Belle2.PyStoreArray('MCParticles')
        if not mcTrackCands:
            return
 
        multiplicity = mcTrackCands.getEntries()
        multiplicity_primaries = multiplicity
 
        # measure the charge asymmetry
        n_matched_plus = 0
        n_matched_minus = 0
 
        for mcTrackCand in mcTrackCands:
            is_matched = trackMatchLookUp.isAnyChargeMatchedMCRecoTrack(mcTrackCand)
 
            relatedPRtrackCand = trackMatchLookUp.getRelatedPRRecoTrack(mcTrackCand)
            if relatedPRtrackCand:
                is_chargeMatched = trackMatchLookUp.isChargeMatched(relatedPRtrackCand)
            else:
                is_chargeMatched = False
 
            if is_chargeMatched:
                if mcTrackCand.getChargeSeed() > 0:
                    n_matched_plus += 1
                else:
                    n_matched_minus += 1
 
            hit_efficiency = trackMatchLookUp.getRelatedEfficiency(mcTrackCand)
            if math.isnan(hit_efficiency):
                hit_efficiency = 0
 
            mcParticle = trackMatchLookUp.getRelatedMCParticle(mcTrackCand)
            mcHelix = getHelixFromMCParticle(mcParticle)
 
            # fill the FilterProperties will all properties on this track
            # gathered so far
            filterProperties = FilterProperties(mcParticle=mcParticle,
                                                mcParticles=mcParticles)
 
            if not self.track_filter_object.doesMcPass(filterProperties):
                continue
 
            momentum = mcParticle.getMomentum()
            pt = momentum.Rho()
            tan_lambda = np.divide(1.0, math.tan(momentum.Theta()))  # Avoid zero division exception
            d0 = mcHelix.getD0()
            det_hit_ids = get_det_hit_ids(mcTrackCand)
            ndf = calc_ndf_from_det_hit_ids(det_hit_ids)
 
            self.mc_matches.append(is_matched)
            self.mc_charge_matches.append(is_chargeMatched and is_matched)
            self.mc_primaries.append(is_primary(mcParticle))
            self.mc_hit_efficiencies.append(hit_efficiency)
            self.mc_pts.append(pt)
            self.mc_d0s.append(d0)
            self.mc_tan_lambdas.append(tan_lambda)
            self.mc_multiplicities.append(multiplicity)
            self.mc_phi.append(momentum.Phi())
            self.mc_ndf.append(ndf)
            if not is_primary(mcParticle):
                multiplicity_primaries -= 1
 
        charge_asymmetry = (n_matched_plus - n_matched_minus)/(n_matched_plus +
                                                               n_matched_minus) if (n_matched_plus + n_matched_minus) != 0 else 0
        for mcTrackCand in mcTrackCands:
            if is_primary(mcParticle):
                self.mc_charge_asymmetry.append(charge_asymmetry)
                self.mc_charge_asymmetry_weights.append(1./multiplicity_primaries)
            else:
                self.mc_charge_asymmetry.append(0)
                self.mc_charge_asymmetry_weights.append(0)
 

examine_pr_tracks()

def examine_pr_tracks (   self )

inherited

Looks at the individual pattern reconstructed tracks and store information about them

Definition at line 265 of file module.py.

initialize()

def initialize

(

self

)

Receive signal at the start of event processing

Reimplemented from TrackingValidationModule.

Definition at line 88 of file hit_module.py.

profiles_by_mc_parameters()

def profiles_by_mc_parameters (   self,   xs,   quantity_name,   unit = None,   parameter_names = [             'd_0',             'p_t',             'tan_lambda',             'multiplicity',             'phi',             'ndf',         ],   make_hist = True,   weights = None,   is_asymmetry = False  )

inherited

Create profile histograms by MC-track parameters

Definition at line 710 of file module.py.

profiles_by_parameters_base()

def profiles_by_parameters_base (   self,   xs,   quantity_name,   parameter_names,   profile_parameters,   unit,   make_hist,   weights = None,   is_asymmetry = False  )

inherited

Create profile histograms for generic parameters

Definition at line 782 of file module.py.

profiles_by_pr_parameters()

def profiles_by_pr_parameters (   self,   xs,   quantity_name,   unit = None,   parameter_names = ['Seed_p_t', 'Seed tan #lambda', 'Seed #phi'],   make_hist = True  )

inherited

Create profile histograms by PR-track parameters

Definition at line 754 of file module.py.

terminate()

def terminate

(

self

)

Receive signal at the end of event processing

Reimplemented from TrackingValidationModule.

Definition at line 308 of file hit_module.py.

Member Data Documentation

pr_clones_and_matches

pr_clones_and_matches

inherited

Use deques in favour of lists to prevent repeated memory allocation of cost O(n)

list of PR-track clones and matches

Definition at line 187 of file module.py.

referenceFileName

referenceFileName

inherited

If this variable is set the code will open the file with same name as the file created here and will read the binning from the TH1/TProfile with same name as the one created here.

If you do not want this feature either remove the corresponding root files from the validation directory (this will trigger the default behaviour) or set the environmental variable DO_NOT_READ_BINNING

Definition at line 169 of file module.py.

---

The documentation for this class was generated from the following file:

• tracking/scripts/tracking/validation/hit_module.py