New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add mask to keep track of all algos where a track is reconstructed (and also move detachedTripletStep right after initialStep in priority order where it should be) Class TkMCQuality is also modified to produce a vector of "qualities" instead of writing a new track collection with a fake quality bit set. #9073
Conversation
A new Pull Request was created by @VinInn (Vincenzo Innocente) for CMSSW_7_5_X. Add mask to keep track all algos where a track is reconstructed It involves the following packages: DataFormats/GsfTrackReco @cmsbuild, @cvuosalo, @nclopezo, @slava77 can you please review it and eventually sign? Thanks. |
|
||
|
||
#endif | ||
#include "DataFormats/TrackReco/interface/trackAlgoPriorityOrder.h" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is the new file missing perhaps?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
was missing indeed...
@cmsbuild please test |
This pull request is fully signed and it will be integrated in one of the next CMSSW_7_5_X IBs unless changes (tests are also fine). This pull request requires discussion in the ORP meeting before it's merged. @davidlange6, @nclopezo, @smuzaffar |
+1 |
Add mask to keep track of all algos where a track is reconstructed (and also move detachedTripletStep right after initialStep in priority order where it should be) Class TkMCQuality is also modified to produce a vector of "qualities" instead of writing a new track collection with a fake quality bit set.
Added a bitset that keep track of all algos a track has been "reconstructed" according to the criteria of the TrackListMerger.
I took the opportunity to fix the algo-priority-order to reflect the actual order in this release.
trackAlgoPriorityOrder.h has been moved to a public header so that it is available also at stages of reconstruction and analysis other than FinalSelectors.
I also modified class TkMCQuality to make it producing a vector of "qualities" instead of writing a new track collection with a fake quality bit set.
All this can be used (among others) to measure the reco efficiency using tag&probe method.
In particular for muon, Iter10 "muonSeededStepOutIn" can be used as a probe to check if the track were found previously.
for instance these are the results on a sample of 2000 boosted Jpsi (particle gun) superimposed to 13TeV pileup (average 35, 25ns)
the first table shows the percentage of reconstructed muon for each couple of algorithms.
The following histograms shows the final assignment to the track as well as which algo found first:
those found first by iter10 are tracker-only inefficiency.
the purple line shows the tracks found only by one algo.
A different view is shown in the second histogram that count how many algo a track has been found in early-iterations. The number of track not found by early-iterations are again the tracker-only inefficiency.
this method of course does not account for tracker-detector inefficiency.
for that standard tag&probe will provide an answer.