Skip to content

Latest commit

 

History

History
70 lines (66 loc) · 3.21 KB

tools.md

File metadata and controls

70 lines (66 loc) · 3.21 KB
Raw

Using tools in Bender

Usage of tools and services in Bender is rather trivial

  1. One aquires tools and services usin the method tool and svc (that are very smilar to thir C++ counterpartners GaudiAlgorithm::tool and GaudiAlgorithm::service). Typically tools and services are aquired in initialize method, e.g.
class TupleTools(Algo):
    """Demonstration how to use TupleTools with Bender"""
    def initialize ( self ) :
        sc = Algo.initialize ( self ) ## initialize the base class
        if sc.isFailure() : return sc        
        IPTT = cpp.IParticleTupleTool
        IETT = cpp.IEventTupleTool        
        self.t1 = self.tool( IPTT,'TupleToolPid'      , parent = self ) ## <--- HERE!
        if not self.t1 : return FAILURE         
        self.t2 = self.tool( IPTT,'TupleToolGeometry' , parent = self ) ## <--- HERE!
        if not self.t2 : return FAILURE
        self.e1 = self.tool( IETT,'TupleToolBeamSpot' , parent = self ) ## <--- HERE!
        if not self.e1 : return FAILURE         
        return SUCCESS
  1. Then tools and services can be direcly used in analyse-method:
## the main 'analysis' method 
def analyse( self ) :   ## IMPORTANT! 
    ...        
    tup = self.nTuple('MyTuple')    
    for b in particles :            
        psi = b ( 1 ) ## the first daughter: J/psi             
        ## Particle Tuple Tool
        sc = self.t1.fill ( b , psi , 'psi_' , tup )
        if sc.isFailure() : return sc            
        ## Particle Tuple Tool 
        sc = self.t2.fill ( b , b   , 'b_'   , tup )
        if sc.isFailure() : return sc                        
        ## Event Tuple Tool 
        sc = self.e1.fill ( tup )
        if sc.isFailure() : return sc                        
        tup.write()

Note that many standard tools and standard services are direclty accessible via the base class DVAlgorithm, e.g.

vx_fitter = self.vertexFitter() 
dcalc     = self.distanceCalculator()
pp        = self.ppSvc()  ##  particle property service

Using tuple tools in Bender

{% challenge "Challenge for the fans of tuple tools use" %}

  1. Add set of known tuple tools into your example
    • (Do not forget to instrument the initialize method to aquire tools)
  2. Run it and observe new varianles into n-tuple.
    • Do their names correspond to your expectations? {% solution "Solution" %} The complete module is accessible here The structure of n-tuple is:
In [7]: from Ostap.Logger import multicolumn
In [8]: f = ROOT.TFile('TupleTools.root','read')
In [9]: t = f.TupleTools.MyTuple
In [10]: print multicolumn ( t.branches() ) 
BeamX              b__ENDVERTEX_COV_  b__ENDVERTEX_Y     b__FDCHI2_OWNPV    b__OWNPV_CHI2      b__OWNPV_XERR      b__OWNPV_ZERR    
BeamY              b__ENDVERTEX_NDOF  b__ENDVERTEX_YERR  b__FD_OWNPV        b__OWNPV_COV_      b__OWNPV_Y         psi__ID          
b__DIRA_OWNPV      b__ENDVERTEX_X     b__ENDVERTEX_Z     b__IPCHI2_OWNPV    b__OWNPV_NDOF      b__OWNPV_YERR    
b__ENDVERTEX_CHI2  b__ENDVERTEX_XERR  b__ENDVERTEX_ZERR  b__IP_OWNPV        b__OWNPV_X         b__OWNPV_Z

{% endchallenge %}