Updates to precision timing integration in reconstruction and miniaod #20338
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -4,50 +4,102 @@ | |
| #include "TrackingTools/TransientTrack/interface/TransientTrackBuilder.h" | ||
| #include "DataFormats/Math/interface/deltaR.h" | ||
| #include "TrackingTools/IPTools/interface/IPTools.h" | ||
| #include "FWCore/Utilities/interface/isFinite.h" | ||
|
|
||
|
|
||
| std::pair<int,PrimaryVertexAssignment::Quality> | ||
| PrimaryVertexAssignment::chargedHadronVertex( const reco::VertexCollection& vertices, | ||
| const reco::TrackRef& trackRef, | ||
| const reco::Track* track, | ||
| float time, | ||
| float timeReso, // <0 if timing not available for this object | ||
| const edm::View<reco::Candidate>& jets, | ||
| const TransientTrackBuilder& builder) const { | ||
|
|
||
| typedef reco::VertexCollection::const_iterator IV; | ||
| typedef reco::Vertex::trackRef_iterator IT; | ||
|
|
||
| int iVertex = -1; | ||
| size_t index=0; | ||
| float bestweight=0; | ||
| for( auto const & vtx : vertices) { | ||
| float w = vtx.trackWeight(trackRef); | ||
| if (w > bestweight){ | ||
| bestweight=w; | ||
| iVertex=index; | ||
| } | ||
| index++; | ||
| } | ||
|
|
||
| bool useTime = useTiming_; | ||
| if (edm::isNotFinite(time) || timeReso<1e-6) { | ||
| useTime = false; | ||
| time = 0.; | ||
| timeReso = -1.; | ||
| } | ||
|
|
||
| if (preferHighRanked_) { | ||
| for(IV iv=vertices.begin(); iv!=vertices.end(); ++iv) { | ||
| int ivtx = iv - vertices.begin(); | ||
| if (iVertex == ivtx) return std::pair<int,PrimaryVertexAssignment::Quality>(ivtx,PrimaryVertexAssignment::UsedInFit); | ||
|
|
||
| double dz = std::abs(track->dz(iv->position())); | ||
| double dt = std::abs(time-iv->t()); | ||
|
|
||
| bool useTimeVtx = useTime && iv->tError()>0.; | ||
|
|
||
| if ((dz < maxDzForPrimaryAssignment_ or dz/track->dzError() < maxDzSigForPrimaryAssignment_ ) and (!useTimeVtx or dt/timeReso < maxDtSigForPrimaryAssignment_)) { | ||
| return std::pair<int,PrimaryVertexAssignment::Quality>(ivtx,PrimaryVertexAssignment::PrimaryDz); | ||
| } | ||
| } | ||
| } | ||
|
|
||
| if(iVertex >= 0 ) return std::pair<int,PrimaryVertexAssignment::Quality>(iVertex,PrimaryVertexAssignment::UsedInFit); | ||
|
|
||
| double distmin = std::numeric_limits<double>::max(); | ||
| double dzmin = std::numeric_limits<double>::max(); | ||
| double dtmin = std::numeric_limits<double>::max(); | ||
| int vtxIdMinDist = -1; | ||
|
There was a problem hiding this comment. vtxIdMinSignif may be a more intuitive name. There was a problem hiding this comment. This is vestigial from the pre-timing version of the algorithm. Can we leave it for future cleanup? |
||
| for(IV iv=vertices.begin(); iv!=vertices.end(); ++iv) { | ||
| double dz = std::abs(track->dz(iv->position())); | ||
| double dt = std::abs(time-iv->t()); | ||
|
|
||
| double dzsig = dz/track->dzError(); | ||
| double dist = dzsig*dzsig; | ||
|
There was a problem hiding this comment. why is the vertex uncertainty not a part of the computation? There was a problem hiding this comment. Yes, that is the reason. Actually I do recall now that the vertex uncertainty IS used in once place later in the file. Let me double check what the reason was. |
||
|
|
||
| bool useTimeVtx = useTime && iv->tError()>0.; | ||
| if (useTime && !useTimeVtx) { | ||
| dt = timeReso; | ||
| } | ||
|
|
||
| if (useTime) { | ||
| double dtsig = dt/timeReso; | ||
|
|
||
| dist += dtsig*dtsig; | ||
| } | ||
| if(dist<distmin) { | ||
| distmin = dist; | ||
| dzmin = dz; | ||
| dtmin = dt; | ||
| vtxIdMinDist = iv-vertices.begin(); | ||
| } | ||
| } | ||
|
|
||
| // first use "closest in Z" with tight cuts (targetting primary particles) | ||
| const float add_cov = vtxIdMinDist >= 0 ? vertices[vtxIdMinDist].covariance(2,2) : 0.f; | ||
| const float dzE=sqrt(track->dzError()*track->dzError()+add_cov); | ||
| if(vtxIdMinDist>=0 and | ||
| (dzmin < maxDzForPrimaryAssignment_ and dzmin/dzE < maxDzSigForPrimaryAssignment_ and track->dzError()<maxDzErrorForPrimaryAssignment_) and | ||
|
There was a problem hiding this comment. There is in fact a small issue which has been introduced here when forward porting this code to 93x and resolving merge conflicts with 9d480b8 It would not affect the non-timing vertices, but could have a small impact on the sorting of the 4d vertices. I'll commit a fix momentarily. There was a problem hiding this comment. Sorry, that's not true, I was misreading/misremembering what this code does. Everything which is here is fine, with indeed a small inconsistency between the handling of z and t, since the vertex error is used here for z, but not for t. I would prefer to leave it like this for now, also because we need to study more closely the pulls on the time errors of the 4d vertices. |
||
| (!useTime or dtmin/timeReso < maxDtSigForPrimaryAssignment_) ) | ||
| { | ||
| iVertex=vtxIdMinDist; | ||
| } | ||
|
|
||
| if(iVertex >= 0 ) return std::pair<int,PrimaryVertexAssignment::Quality>(iVertex,PrimaryVertexAssignment::PrimaryDz); | ||
|
|
||
|
|
||
|
|
||
| // if track not assigned yet, it could be a b-decay secondary , use jet axis dist criterion | ||
| // first find the closest jet within maxJetDeltaR_ | ||
| int jetIdx = -1; | ||
|
|
@@ -94,14 +146,14 @@ PrimaryVertexAssignment::chargedHadronVertex( const reco::VertexCollection& vert | |
|
|
||
| // if the track is not compatible with other PVs but is compatible with the BeamSpot, we may simply have not reco'ed the PV! | ||
| // we still point it to the closest in Z, but flag it as possible orphan-primary | ||
| if(!vertices.empty() && std::abs(track->dxy(vertices[0].position()))<maxDxyForNotReconstructedPrimary_ && std::abs(track->dxy(vertices[0].position())/track->dxyError())<maxDxySigForNotReconstructedPrimary_) | ||
| return std::pair<int,PrimaryVertexAssignment::Quality>(vtxIdMinDist,PrimaryVertexAssignment::NotReconstructedPrimary); | ||
|
|
||
| //FIXME: here we could better handle V0s and NucInt | ||
|
|
||
| // all other tracks could be non-B secondaries and we just attach them with closest Z | ||
| if(vtxIdMinDist>=0) | ||
| return std::pair<int,PrimaryVertexAssignment::Quality>(vtxIdMinDist,PrimaryVertexAssignment::OtherDz); | ||
| //If for some reason even the dz failed (when?) we consider the track not assigned | ||
| return std::pair<int,PrimaryVertexAssignment::Quality>(-1,PrimaryVertexAssignment::Unassigned); | ||
| } | ||
|
|
||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
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.
why is the vertex error not a part of the significance computation here (it is in other places in this file)?
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.
the vertex supposedly has a time resolution at this point.
Shouldn't it be used as well in the significance computation?
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.
The vertex time error is only checked if it's >0 (ie to check that the vertex has valid timing info.)
The numerical value is intentionally not used in order to avoid preferring to assign tracks to vertices with large time uncertainty (and the logic is coherent for z and t).
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.
(where the dz significance is used later in the code including the vertex error, it will not cause poor vertices to be favoured, because it is only checked against the closest vertex)
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.
the reason to add the vtx z error was to "regularize" the selection for tracks with very small dzError. IIRC, values down to 5 um are possible which can simply kill a good track with respect to still rather reasonable vertices.