Acts::AmbiguityResolutionML< AmbiguityNetwork > Class Template Reference

Generic implementation of the machine learning ambiguity resolution Contains method for data preparations. More...

#include <Acts/AmbiguityResolution/AmbiguityResolutionML.hpp>

Classes
struct	Config

Public Member Functions
	AmbiguityResolutionML (const Config &cfg, std::unique_ptr< const Logger > logger=getDefaultLogger("AmbiguityResolutionML", Logging::INFO))
	Construct the ambiguity resolution algorithm.
template<TrackContainerFrontend track_container_t, typename source_link_hash_t, typename source_link_equality_t>
std::multimap< int, std::pair< std::size_t, std::vector< std::size_t > > >	mapTrackHits (const track_container_t &tracks, const source_link_hash_t &sourceLinkHash, const source_link_equality_t &sourceLinkEquality) const
	Associate the hits to the tracks.
template<TrackContainerFrontend track_container_t>
std::vector< std::size_t >	solveAmbiguity (std::unordered_map< std::size_t, std::vector< std::size_t > > &clusters, const track_container_t &tracks) const
	Select the track associated with each cluster.

Detailed Description

template<AmbiguityNetworkConcept AmbiguityNetwork>
class Acts::AmbiguityResolutionML< AmbiguityNetwork >

Generic implementation of the machine learning ambiguity resolution Contains method for data preparations.

Constructor & Destructor Documentation

AmbiguityResolutionML()

template<AmbiguityNetworkConcept AmbiguityNetwork>

Acts::AmbiguityResolutionML< AmbiguityNetwork >::AmbiguityResolutionML ( const Config & cfg, std::unique_ptr< const Logger > logger = getDefaultLogger("AmbiguityResolutionML< AmbiguityNetwork >",                                                      Logging::INFO) )

explicit

Construct the ambiguity resolution algorithm.

Parameters

cfg	is the algorithm configuration
logger	is the logging instance

Member Function Documentation

mapTrackHits()

template<AmbiguityNetworkConcept AmbiguityNetwork>

template<TrackContainerFrontend track_container_t, typename source_link_hash_t, typename source_link_equality_t>

std::multimap< int, std::pair< std::size_t, std::vector< std::size_t > > > Acts::AmbiguityResolutionML< AmbiguityNetwork >::mapTrackHits	(	const track_container_t &	tracks,
		const source_link_hash_t &	sourceLinkHash,
		const source_link_equality_t &	sourceLinkEquality ) const

Associate the hits to the tracks.

This algorithm performs the mapping of hits ID to track ID. Our final goal is too loop over all the tracks (and their associated hits) by order of decreasing number hits for this we use a multimap where the key is the number of hits as this will automatically perform the sorting.

Parameters

tracks	is the input track container
sourceLinkHash	is the hash function for the source link, will be used to associate to tracks
sourceLinkEquality	is the equality function for the source link used used to associated hits to tracks

Returns

an ordered list containing pairs of track ID and associated measurement ID

solveAmbiguity()

template<AmbiguityNetworkConcept AmbiguityNetwork>

template<TrackContainerFrontend track_container_t>

std::vector< std::size_t > Acts::AmbiguityResolutionML< AmbiguityNetwork >::solveAmbiguity	(	std::unordered_map< std::size_t, std::vector< std::size_t > > &	clusters,
		const track_container_t &	tracks ) const

Select the track associated with each cluster.

In this algorithm the call the neural network to score the tracks and then select the track with the highest score in each cluster

Parameters

clusters	is a map of clusters, each cluster correspond to a vector of track ID
tracks	is the input track container

Returns

a vector of trackID corresponding tho the good tracks