#include <DivisiveClusterizer1D.h>

Inheritance diagram for DivisiveClusterizer1D< T >:

Public Member Functions
virtual DivisiveClusterizer1D *	clone () const

	DivisiveClusterizer1D (float zoffset=5., int ntkmin=5, bool useError=true, float zsep=0.05, bool wei=true)

std::pair< std::vector < Cluster1D< T > >, std::vector< const T * > >	operator() (const std::vector< Cluster1D< T > > &) const

	~DivisiveClusterizer1D ()

Public Member Functions inherited from Clusterizer1D< T >
virtual	~Clusterizer1D ()

Private Member Functions
void	findCandidates (const std::vector< Cluster1D< T > > &, std::vector< Cluster1D< T > > &, std::vector< Cluster1D< T > > &) const

void	insertTracks (std::vector< Cluster1D< T > > &, std::vector< Cluster1D< T > > &) const

std::vector< Cluster1D< T > >	makeCluster1Ds (std::vector< Cluster1D< T > > &, std::vector< Cluster1D< T > > &) const

Cluster1D< T >	mergeCluster1Ds (std::vector< Cluster1D< T > > &) const

std::vector< const T * >	takeTracks (const std::vector< Cluster1D< T > > &) const

Private Attributes
Cluster1DCleaner< T > *	theCleaner

Cluster1DMerger< T > *	theMerger

unsigned	theNTkMin

bool	theUseError

bool	theWei

float	theZOffSet

float	theZSeparation

Detailed Description

template<class T>
class DivisiveClusterizer1D< T >

Find the modes with a simple divisive method.

Definition at line 14 of file DivisiveClusterizer1D.h.

Constructor & Destructor Documentation

template<class T >

DivisiveClusterizer1D< T >::DivisiveClusterizer1D	(	float	zoffset = `5.`,
		int	ntkmin = `5`,
		bool	useError = `true`,
		float	zsep = `0.05`,
		bool	wei = `true`
	)

Parameters

zoffset	maximum distance between track position and position of its cluster (depending on useError its either weighted or physical distance)
ntkmin	Minimum number of tracks required to form a cluster.
useError	physical distances or weighted distances.
zsep	Maximum distance between two adjacent tracks that belong to the same initial cluster.
wei	Compute the cluster "center" with an unweighted or a weighted average of the tracks. Weighted means weighted with the error of the data point.

Definition at line 67 of file DivisiveClusterizer1D.h.

References DivisiveClusterizer1D< T >::theCleaner, DivisiveClusterizer1D< T >::theMerger, DivisiveClusterizer1D< T >::theUseError, and DivisiveClusterizer1D< T >::theZOffSet.

         :theZOffSet(zoffset), theZSeparation(zsep), theNTkMin(ntkmin),
         theWei(wei), theUseError(useError)
 {
     //  theDiscardedTracks.clear();
     // theTotalDiscardedTracks.clear();
     //  theCluster1Ds.clear();
     TrivialWeightEstimator<T> weightEstimator;
     theMerger = new Cluster1DMerger<T>( weightEstimator );
     theCleaner = new Cluster1DCleaner<T>(theZOffSet,theUseError);
 }

template<class T >

DivisiveClusterizer1D< T >::~DivisiveClusterizer1D ( )

Definition at line 83 of file DivisiveClusterizer1D.h.

 {
     delete theMerger;
     delete theCleaner;
 }

Member Function Documentation

template<class T >

DivisiveClusterizer1D< T > * DivisiveClusterizer1D< T >::clone ( void ) const

virtual

Implements Clusterizer1D< T >.

Definition at line 102 of file DivisiveClusterizer1D.h.

 {
     return new DivisiveClusterizer1D<T> ( * this );
 }

template<class T >

void DivisiveClusterizer1D< T >::findCandidates	(	const std::vector< Cluster1D< T > > &	inputo,
		std::vector< Cluster1D< T > > &	finalCluster1Ds,
		std::vector< Cluster1D< T > > &	totDiscardedTracks
	)		const

private

Definition at line 109 of file DivisiveClusterizer1D.h.

References collect_tpl::input, and python.multivaluedict::sort().

 {
     using namespace Clusterizer1DCommons;
 
     std::vector < Cluster1D<T> > input = inputo;
     std::vector < Cluster1D<T> > discardedTracks;
     if ( input.size() < theNTkMin)
     {
         insertTracks( input, totDiscardedTracks);
         return;
     }
     sort( input.begin(), input.end(), ComparePairs<T>());
     int ncount =0;
     std::vector < Cluster1D<T> > partOfPTracks;
     partOfPTracks.push_back(input.front());
     for( typename std::vector < Cluster1D<T> >::const_iterator ic=(input.begin())+1;
             ic != input.end(); ic++)
     {
         ncount++;
         if ( fabs( (*ic).position().value()-(*(ic-1)).position().value() )
                 < (double) theZSeparation )
         {
             partOfPTracks.push_back((*ic));
         }
         else
         {
             if(partOfPTracks.size() >= theNTkMin)
             {
                 std::vector< Cluster1D<T> > clusters = makeCluster1Ds(partOfPTracks,
                                                      discardedTracks);
                 for (typename std::vector< Cluster1D<T> >::const_iterator iclus=clusters.begin();
                         iclus != clusters.end(); iclus++)
                 {
                     finalCluster1Ds.push_back(*iclus);
                 }
                 insertTracks(discardedTracks,totDiscardedTracks);
             }
             else
             {
                 insertTracks(partOfPTracks,totDiscardedTracks);
             }
             partOfPTracks.clear();
             partOfPTracks.push_back((*ic));
         }
     }
     if (partOfPTracks.size() >= theNTkMin)
     {
         std::vector < Cluster1D<T> > clusters = makeCluster1Ds(partOfPTracks,
                                               discardedTracks);
         for (typename std::vector< Cluster1D<T> >::const_iterator iclus = clusters.begin();
                 iclus != clusters.end(); iclus++)
         {
             finalCluster1Ds.push_back(*iclus);
         }
         insertTracks(discardedTracks,totDiscardedTracks);
     }
     else
     {
         insertTracks(partOfPTracks,totDiscardedTracks);
     }
 
     sort(finalCluster1Ds.begin(), finalCluster1Ds.end(),
          ComparePairs<T>() );
     // reverse(theCluster1Ds.begin(), theCluster1Ds.end());
 
     return;
 }

template<class T >

void DivisiveClusterizer1D< T >::insertTracks	(	std::vector< Cluster1D< T > > &	clusou,
		std::vector< Cluster1D< T > > &	cludest
	)		const

private

Definition at line 233 of file DivisiveClusterizer1D.h.

 {
     if (clusou.size() == 0)
         return;
     for ( typename std::vector< Cluster1D<T> >::const_iterator iclu = clusou.begin();
             iclu != clusou.end(); iclu++)
     {
         cludest.push_back(*iclu);
     }
     /*
     for ( typename std::vector< Cluster1D<T> >::const_iterator iclu = clu.begin(); 
     iclu != clu.end(); iclu++){
       if (total) {
         theTotalDiscardedTracks.push_back(*iclu);
       }else { 
         theDiscardedTracks.push_back(*iclu);
       }
     }
     */
     return;
 }

template<class T >

std::vector< Cluster1D< T > > DivisiveClusterizer1D< T >::makeCluster1Ds	(	std::vector< Cluster1D< T > > &	clusters,
		std::vector< Cluster1D< T > > &	discardedTracks
	)		const

private

Definition at line 181 of file DivisiveClusterizer1D.h.

 {
 
     std::vector < Cluster1D<T> > finalCluster1Ds;
     discardedTracks.clear();
     std::vector<Cluster1D<T> > pvClu0 = clusters;
     std::vector<Cluster1D<T> > pvCluNew = pvClu0;
     bool stop = false;
     while (!stop)
     {
         int nDiscardedAtIteration = 100;
         while ( nDiscardedAtIteration !=0 )
         {
             pvCluNew = theCleaner->clusters(pvClu0);
             std::vector<Cluster1D<T> >
             tracksAtIteration = theCleaner->discardedCluster1Ds();
             nDiscardedAtIteration = tracksAtIteration.size();
             if ( nDiscardedAtIteration!=0 )
             {
                 insertTracks(tracksAtIteration,discardedTracks);
                 pvClu0 = pvCluNew;
             }
         } // while nDiscardedAtIteration
         unsigned ntkclus = pvCluNew.size();
         unsigned ndiscard = discardedTracks.size();
 
         if ( ntkclus >=  theNTkMin )
         {
             //save the cluster
             finalCluster1Ds.push_back( mergeCluster1Ds(pvCluNew) );
             if ( ndiscard >= theNTkMin )
             { //make a new cluster and reset
                 pvClu0 = discardedTracks;
                 discardedTracks.clear();
             }
             else
             { //out of loop
                 stop = true;
             }
         }
         else
         {
             insertTracks(pvCluNew, discardedTracks);
             stop = true;
         }
     }// while stop
     return finalCluster1Ds;
 }

template<class T >

Cluster1D< T > DivisiveClusterizer1D< T >::mergeCluster1Ds ( std::vector< Cluster1D< T > > & clusters ) const

private

Definition at line 276 of file DivisiveClusterizer1D.h.

References query::result.

 {
 
     Cluster1D<T> result = clusters.front();
     for ( typename std::vector< Cluster1D<T> >::iterator iclu = (clusters.begin())+1;
             iclu != clusters.end(); iclu++)
     {
         Cluster1D<T> old = result;
         result = (*theMerger)( old , *iclu );
     }
     return result;
 }

template<class T >

std::pair< std::vector< Cluster1D< T > >, std::vector< const T * > > DivisiveClusterizer1D< T >::operator() ( const std::vector< Cluster1D< T > > & input ) const

virtual

Implements Clusterizer1D< T >.

Definition at line 92 of file DivisiveClusterizer1D.h.

References collect_tpl::input, and convertSQLitetoXML_cfg::output.

 {
     std::vector < Cluster1D<T> > discardedCluster1Ds;
     std::vector < Cluster1D<T> > output;
     findCandidates ( input, output, discardedCluster1Ds );
     return std::pair < std::vector < Cluster1D<T> >, std::vector < const T * > >
            ( output, takeTracks(discardedCluster1Ds) );
 }

template<class T >

std::vector< const T * > DivisiveClusterizer1D< T >::takeTracks ( const std::vector< Cluster1D< T > > & clu ) const

private

Definition at line 258 of file DivisiveClusterizer1D.h.

References i, and testEve_cfg::tracks.

 {
     std::vector<const T* > tracks;
     for ( typename std::vector< Cluster1D<T> >::const_iterator iclu = clu.begin();
             iclu != clu.end(); iclu++)
     {
         std::vector < const T * > clutks = iclu->tracks();
         for ( typename std::vector< const T * >::const_iterator i=clutks.begin();
                 i!=clutks.end() ; ++i )
         {
             tracks.push_back ( *i );
         }
     }
     return tracks;
 }