Skip to content

Latest commit

 

History

History
42 lines (37 loc) · 2.99 KB

mcmatch.md

File metadata and controls

42 lines (37 loc) · 2.99 KB
Raw

Match reco with MC-truth

Matching of recontructed candidates with MC-truth is performed using MCTRUTH functors, that allows to answer on the basic question: does this MC-particle makes the contribution to this reconstructed particle?. Note that this question is different from e.g. what is MC-truth for this reconstructed particle?. These are different questions, and therefore one should not misinterpret the answers. For more details see the chapter 15 in LHCb-2004-023.

There are helper methods mcTruth, that are needed to create the functor MCTRUTH

mcK = self.mcselect ( 'mcK' , '[Beauty ==>  J/psi(1S) ^K+ ^K- pi+ pi-]CC' ) ## get true MC-kaons 
trueK = MCTRUTH ( mcK , mc.mcTruth() ) ## <--- HERE: create MCTRUTH    functor

The created object trueK is LoKi fuctor, that evaluates to True for reconstructed particles, that get the contribution from true MC kaons, selected by the mcselect method, otherwise it evalautes to `False':

reco_kaons = self.select ( ... )
for k in reco_kaons : 
    print ' True(MC-truth matched) kaon?  %s' % trueK ( k )

Since it is an ordinary LoKi functor is could be combined with all other functors, e.g. one can select from the input only MC-truth matcehd kaons:

truth_matched_kaons = self.select ( 'K' ,  ( 'K+' == ABSID ) & trueK )

{% discussion "Could it be inverted?" %} the short answer is yes, there is inverse for MCTRUTH functor. The inverse functor is RCTRUTH, it evalauted for True for any MC particle, that makes contribution to the selected recontructed candidate, and False otherwise.

B = self.select ( 'B' ,  'Beauty -> J/psi(1S)  K+  K- pi+ pi-' ) 
## get kaons from our reconstrcuted B-candidates 
K = self.select ( 'K' , '[Beauty -> J/psi(1S) ^K+ ^K- pi+ pi-' )

recoK = RCTRUTH ( K , self.mcTruth() )

##  get all MC-particles that makes contribution to recontructed Kaon
mc = self.mcselect ( 'mc' , recoK ) 
print mc

Note that the list of found MC-particles here could be rather long (do you remember the basic question?) {% enddiscussion %}

{% challenge "Challenge" %} Try to code some MC-truth match algorithm, that get some MC-decays, some reconstructed decays, and perform MC-truth macth between them. Try to select as an example the ordinary ALLSTREAMS.DST (not Turbo version!). Processing of ALLSTREAMS.DST/Turbo and various kinds of ALLSTREAMS.MDST and MC-Turbo requires a bit different configurtaion steos, that we'll discuss later. For a time being y ou can use e.g. the MC-file '/lhcb/MC/2012/ALLSTREAMS.DST/00033494/0000/00033494_00000013_1.allstreams.dst', that contains the true MC-decays Bs -> J/psi K+ K- pi+ pi- with many intermediate resonances. {% solution "Solution" %} The complete module, that processes the events of [B_s0 ==> J/psi(1S) K+ K- pi+ pi-]CC with very rich structure of intermediate resonances is available here {% endchallenge %}