Enable DNN supercluster in Phase2 HLT #46010

RSalvatico · 2024-09-17T06:19:32Z

PR description:

Enable the usage of the DNN superclustering developed for the endcaps (PR #45333) in the Phase2 HLT reconstruction. In particular, this PR allows us to use:

Superclustering DNN (TracksterLinkingBySuperClusteringDNN)
Superclustering with Mustache inside TICL (TracksterLinkingbySuperClusteringMustache)

using process modifiers.

PR validation:

Successfully run several upgrade workflows such as
29691.203 SingleElectronPt15Eta1p7_2p7_2026D110 + ticl_v5
29691.204 SingleElectronPt15Eta1p7_2p7_2026D110 + ticl_v5_mustache

cmsbuild · 2024-09-17T06:20:00Z

cms-bot internal usage

cmsbuild · 2024-09-17T06:21:19Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-46010/41813

There are other open Pull requests which might conflict with changes you have proposed:
- File Configuration/PyReleaseValidation/python/upgradeWorkflowComponents.py modified in PR(s): Add run3 tracking low pu era #33532, CMSSW Integration of LST #45117, Introduce support for cluster splitting and JetCore iteration in Phase-2 track reconstruction #46001, FastSim premixing workflows and matrix cleanup #46007

cmsbuild · 2024-09-17T06:21:42Z

A new Pull Request was created by @RSalvatico for master.

It involves the following packages:

Configuration/PyReleaseValidation (upgrade, pdmv)
HLTrigger/Configuration (hlt)

@AdrianoDee, @Martin-Grunewald, @cmsbuild, @kskovpen, @miquork, @mmusich, @srimanob, @subirsarkar, @sunilUIET can you please review it and eventually sign? Thanks.
@Martin-Grunewald, @SohamBhattacharya, @VourMa, @fabiocos, @makortel, @missirol, @mmusich, @rovere, @silviodonato, @slomeo this is something you requested to watch as well.
@antoniovilela, @mandrenguyen, @rappoccio, @sextonkennedy you are the release manager for this.

cms-bot commands are listed here

felicepantaleo · 2024-09-17T06:23:01Z

test parameters:

workflow_opts= -w upgrade
workflows = 29888.203,29688.203

felicepantaleo · 2024-09-17T06:23:39Z

@cmsbuild please test

felicepantaleo · 2024-09-17T06:29:48Z

@DebabrataBhowmik @waredjeb @rovere @VourMa fyi

cmsbuild · 2024-09-17T11:34:59Z

+1

Size: This PR adds an extra 68KB to repository
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-eb053e/41547/summary.html
COMMIT: a85bda5
CMSSW: CMSSW_14_2_X_2024-09-16-2300/el8_amd64_gcc12
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/46010/41547/install.sh to create a dev area with all the needed externals and cmssw changes.

DAS Queries: The DAS query tests failed, see the summary page for details.

Comparison Summary

Summary:

You potentially removed 2 lines from the logs
Reco comparison results: 10 differences found in the comparisons
DQMHistoTests: Total files compared: 46
DQMHistoTests: Total histograms compared: 3532200
DQMHistoTests: Total failures: 511
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 3531669
DQMHistoTests: Total skipped: 20
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 45 files compared)
Checked 202 log files, 171 edm output root files, 46 DQM output files
TriggerResults: found differences in 2 / 44 workflows

mmusich · 2024-09-17T11:37:45Z

enable hlt_p2_timing

mmusich

cosmetics

mmusich · 2024-09-17T07:15:13Z

HLTrigger/Configuration/python/HLT_75e33/modules/hltEgammaCandidatesL1Seeded_cfi.py

@@ -5,3 +5,6 @@
    scHybridBarrelProducer = cms.InputTag("hltParticleFlowSuperClusterECALL1Seeded","particleFlowSuperClusterECALBarrel"),
    scIslandEndcapProducer = cms.InputTag("hltParticleFlowSuperClusterHGCalFromTICLL1Seeded")
 )
+
+from Configuration.ProcessModifiers.ticl_v5_cff import ticl_v5
+ticl_v5.toModify(hltEgammaCandidatesL1Seeded, scIslandEndcapProducer = cms.InputTag("hltTiclEGammaSuperClusterProducerL1Seeded"))


missing newline.

mmusich · 2024-09-17T07:15:25Z

HLTrigger/Configuration/python/HLT_75e33/modules/hltEgammaCandidatesUnseeded_cfi.py

@@ -5,3 +5,6 @@
    scHybridBarrelProducer = cms.InputTag("hltParticleFlowSuperClusterECALUnseeded","particleFlowSuperClusterECALBarrel"),
    scIslandEndcapProducer = cms.InputTag("hltParticleFlowSuperClusterHGCalFromTICLUnseeded")
 )
+
+from Configuration.ProcessModifiers.ticl_v5_cff import ticl_v5
+ticl_v5.toModify(hltEgammaCandidatesUnseeded, scIslandEndcapProducer = cms.InputTag("hltTiclEGammaSuperClusterProducerUnseeded"))


missing newline.

mmusich · 2024-09-17T07:15:29Z

HLTrigger/Configuration/python/HLT_75e33/modules/hltTiclTracksterLinksL1Seeded_cfi.py

+    regressionAndPid = cms.bool(True),
+    tfDnnLabel = cms.string('tracksterSelectionTf'),
+    tracksters_collections = cms.VInputTag("hltTiclTrackstersCLUE3DHighL1Seeded", "hltTiclTrackstersPassthrough")
+)


missing newline.

mmusich · 2024-09-17T07:15:35Z

HLTrigger/Configuration/python/HLT_75e33/modules/hltTiclTracksterLinksUnseeded_cfi.py

+    regressionAndPid = cms.bool(True),
+    tfDnnLabel = cms.string('tracksterSelectionTf'),
+    tracksters_collections = cms.VInputTag("hltTiclTrackstersCLUE3DHigh", "hltTiclTrackstersPassthrough")
+)


missing newline.

mmusich · 2024-09-17T07:15:39Z

...guration/python/HLT_75e33/sequences/HLTHgcalTiclPFClusteringForEgammaL1SeededSequence_cfi.py

+                                  ticlSuperClusters=cms.InputTag("hltTiclTracksterLinksSuperclusteringMustacheL1Seeded"),
+                                  ticlTrackstersEM=cms.InputTag("hltTiclTrackstersCLUE3DHighL1Seeded"),
+                                  layerClusters=cms.InputTag("hltHgcalMergeLayerClustersL1Seeded")
+                                  )


missing newline.

mmusich · 2024-09-17T07:15:53Z

...guration/python/HLT_75e33/sequences/HLTHgcalTiclPFClusteringForEgammaUnseededSequence_cfi.py

+(alpaka & ticl_superclustering_mustache_ticl).toModify(hltTiclEGammaSuperClusterProducerUnseeded, 
+                                                       ticlSuperClusters=cms.InputTag("hltTiclTracksterLinksSuperclusteringMustacheUnseeded"),
+                                                       ticlTrackstersEM=cms.InputTag("hltTiclTrackstersCLUE3DHigh"),
+                                                       layerClusters=cms.InputTag("hltHgcalMergeLayerClusters"))


missing newline.

VourMa · 2024-09-17T12:28:44Z

...guration/python/HLT_75e33/sequences/HLTHgcalTiclPFClusteringForEgammaUnseededSequence_cfi.py

+(alpaka & ticl_superclustering_dnn).toReplaceWith(_HgcalLocalRecoUnseededSequence, 
+                    cms.Sequence(
+                        hltHgcalDigis
+                        + hltHGCalUncalibRecHit
+                        + hltHGCalRecHit+hltParticleFlowRecHitHGC
+                        + hltHgcalSoARecHitsProducer
+                        + hltHgcalSoARecHitsLayerClustersProducer
+                        + hltHgcalSoALayerClustersProducer
+                        + hltHgCalLayerClustersFromSoAProducer
+                        + hltHgcalLayerClustersHSci
+                        + hltHgcalLayerClustersHSi
+                        + hltHgcalMergeLayerClusters
+                        ) 
+)


How does this differ from

cmssw/HLTrigger/Configuration/python/HLT_75e33/sequences/HLTHgcalTiclPFClusteringForEgammaUnseededSequence_cfi.py

Lines 30 to 45 in a85bda5

# Alpaka

from Configuration.ProcessModifiers.alpaka_cff import alpaka

alpaka.toReplaceWith(_HgcalLocalRecoUnseededSequence,

cms.Sequence(

hltHgcalDigis

+ hltHGCalUncalibRecHit

+ hltHGCalRecHit+hltParticleFlowRecHitHGC

+ hltHgcalSoARecHitsProducer

+ hltHgcalSoARecHitsLayerClustersProducer

+ hltHgcalSoALayerClustersProducer

+ hltHgCalLayerClustersFromSoAProducer

+ hltHgcalLayerClustersHSci

+ hltHgcalLayerClustersHSi

+ hltHgcalMergeLayerClusters

)

)

?

Don't the aforementioned lines cover the heterogeneous workflow more in general? Maybe I am missing how the ticl_superclustering_dnn modifier changes anything here.

No, I think you're right. In fact all the (alpaka & ticl_superclustering) modifiers are pointless, as alpaka and ticl_superclustering modify independent sequences. Thanks

VourMa · 2024-09-17T12:29:23Z

...guration/python/HLT_75e33/sequences/HLTHgcalTiclPFClusteringForEgammaUnseededSequence_cfi.py

+(alpaka & ticl_superclustering_mustache_ticl).toReplaceWith(_HgcalLocalRecoUnseededSequence, 
+                    cms.Sequence(
+                        hltHgcalDigis
+                        + hltHGCalUncalibRecHit
+                        + hltHGCalRecHit+hltParticleFlowRecHitHGC
+                        + hltHgcalSoARecHitsProducer
+                        + hltHgcalSoARecHitsLayerClustersProducer
+                        + hltHgcalSoALayerClustersProducer
+                        + hltHgCalLayerClustersFromSoAProducer
+                        + hltHgcalLayerClustersHSci
+                        + hltHgcalLayerClustersHSi
+                        + hltHgcalMergeLayerClusters
+                        ) 
+)


Same comment as above.

VourMa · 2024-09-17T12:40:23Z

...guration/python/HLT_75e33/sequences/HLTHgcalTiclPFClusteringForEgammaUnseededSequence_cfi.py

+(alpaka & ticl_superclustering_dnn).toModify(hltHgcalMergeLayerClusters,
+                                             layerClustersEE = cms.InputTag("hltHgCalLayerClustersFromSoAProducer"),
+                                             time_layerclustersEE = cms.InputTag("hltHgCalLayerClustersFromSoAProducer", "timeLayerCluster"))


Similar comment as above when compared to lines

cmssw/HLTrigger/Configuration/python/HLT_75e33/sequences/HLTHgcalTiclPFClusteringForEgammaUnseededSequence_cfi.py

Lines 46 to 48 in a85bda5

alpaka.toModify(hltHgcalMergeLayerClusters,

layerClustersEE = cms.InputTag("hltHgCalLayerClustersFromSoAProducer"),

time_layerclustersEE = cms.InputTag("hltHgCalLayerClustersFromSoAProducer", "timeLayerCluster"))

Similarly applied to

cmssw/HLTrigger/Configuration/python/HLT_75e33/sequences/HLTHgcalTiclPFClusteringForEgammaUnseededSequence_cfi.py

Lines 142 to 144 in a85bda5

(alpaka & ticl_superclustering_mustache_ticl).toModify(hltHgcalMergeLayerClusters,

layerClustersEE = cms.InputTag("hltHgCalLayerClustersFromSoAProducer"),

time_layerclustersEE = cms.InputTag("hltHgCalLayerClustersFromSoAProducer", "timeLayerCluster"))

In general the "# Alpaka & DNN" part seems to be duplicating modifications from above, potentially making the maintenance of the code harder. Maybe try to simplify - unless there is some reason for the duplication which I failed to see (potentially having to do with the order the modifiers are applied)?

Yes indeed, I'll clean up.

mmusich · 2024-10-09T07:27:17Z

test parameters:

workflow_opts= -w upgrade
workflows = 29691.203, 29691.204
enable = hlt_p2_integration

mmusich · 2024-10-09T07:27:24Z

@cmsbuild, please test

cmsbuild · 2024-10-09T11:26:50Z

+1

Size: This PR adds an extra 28KB to repository
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-eb053e/42065/summary.html
COMMIT: 0136414
CMSSW: CMSSW_14_2_X_2024-10-08-2300/el8_amd64_gcc12
Additional Tests: HLT_P2_INTEGRATION
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/46010/42065/install.sh to create a dev area with all the needed externals and cmssw changes.

DAS Queries: The DAS query tests failed, see the summary page for details.

Comparison Summary

Summary:

You potentially removed 3 lines from the logs
ROOTFileChecks: Some differences in event products or their sizes found
Reco comparison results: 129 differences found in the comparisons
DQMHistoTests: Total files compared: 46
DQMHistoTests: Total histograms compared: 3532042
DQMHistoTests: Total failures: 5237
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 3526785
DQMHistoTests: Total skipped: 20
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 45 files compared)
Checked 203 log files, 171 edm output root files, 46 DQM output files
TriggerResults: found differences in 2 / 44 workflows

mmusich · 2024-10-17T09:48:48Z

+hlt

for RSalvatico@0136414
PR according to description and follow-up review
phase-2 HLT integration tests are passed
private tests using the modifiers passed: Enable DNN supercluster in Phase2 HLT #46010 (comment)
no sizeable increase of timing: Enable DNN supercluster in Phase2 HLT #46010 (comment)
for a discussion of the changes in the DQM, see Enable DNN supercluster in Phase2 HLT #46010 (comment) and following

jfernan2 · 2024-10-17T17:17:46Z

+4

Moanwar · 2024-10-17T17:19:25Z

+Upgrade

AdrianoDee · 2024-10-17T20:02:53Z

+pdmv

RSalvatico · 2024-10-18T07:42:09Z

+4

Hi @jfernan2, one signature would be enough for me :)

jfernan2 · 2024-10-18T11:00:27Z

+1

cmsbuild · 2024-10-18T11:00:54Z

This pull request is fully signed and it will be integrated in one of the next master IBs (tests are also fine). This pull request will now be reviewed by the release team before it's merged. @mandrenguyen, @sextonkennedy, @rappoccio, @antoniovilela (and backports should be raised in the release meeting by the corresponding L2)

mandrenguyen · 2024-10-18T12:36:36Z

+1

cmsbuild added this to the CMSSW_14_2_X milestone Sep 17, 2024

cmsbuild added hlt-pending pending-signatures tests-pending orp-pending pdmv-pending upgrade-pending code-checks-pending labels Sep 17, 2024

cmsbuild added code-checks-approved and removed code-checks-pending labels Sep 17, 2024

cmsbuild added tests-started and removed tests-pending labels Sep 17, 2024

cmsbuild added tests-approved and removed tests-started labels Sep 17, 2024

mmusich reviewed Sep 17, 2024

View reviewed changes

VourMa reviewed Sep 17, 2024

View reviewed changes

cmsbuild added tests-pending code-checks-pending and removed tests-approved code-checks-approved labels Sep 17, 2024

cmsbuild added the tests-approved label Oct 8, 2024

cmsbuild added tests-started and removed tests-approved labels Oct 9, 2024

cmsbuild added tests-approved and removed tests-started labels Oct 9, 2024

cmsbuild added hlt-approved and removed hlt-pending labels Oct 17, 2024

cmsbuild added upgrade-approved and removed upgrade-pending labels Oct 17, 2024

cmsbuild added pdmv-approved and removed pdmv-pending labels Oct 17, 2024

cmsbuild added reconstruction-approved fully-signed and removed reconstruction-pending pending-signatures labels Oct 18, 2024

cmsbuild added orp-approved and removed orp-pending labels Oct 18, 2024

cmsbuild merged commit eee03e9 into cms-sw:master Oct 18, 2024
13 checks passed

RSalvatico deleted the dnnHLT branch October 19, 2024 19:03

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Enable DNN supercluster in Phase2 HLT #46010

Enable DNN supercluster in Phase2 HLT #46010

RSalvatico commented Sep 17, 2024

cmsbuild commented Sep 17, 2024 •

edited

Loading

cmsbuild commented Sep 17, 2024

cmsbuild commented Sep 17, 2024

felicepantaleo commented Sep 17, 2024

felicepantaleo commented Sep 17, 2024

felicepantaleo commented Sep 17, 2024

cmsbuild commented Sep 17, 2024

mmusich commented Sep 17, 2024

mmusich left a comment

mmusich Sep 17, 2024

mmusich Sep 17, 2024

mmusich Sep 17, 2024

mmusich Sep 17, 2024

mmusich Sep 17, 2024

mmusich Sep 17, 2024

VourMa Sep 17, 2024

RSalvatico Sep 17, 2024 •

edited

Loading

VourMa Sep 17, 2024

VourMa Sep 17, 2024

RSalvatico Sep 17, 2024

mmusich commented Oct 9, 2024

mmusich commented Oct 9, 2024

cmsbuild commented Oct 9, 2024

mmusich commented Oct 17, 2024

jfernan2 commented Oct 17, 2024

Moanwar commented Oct 17, 2024

AdrianoDee commented Oct 17, 2024

RSalvatico commented Oct 18, 2024

jfernan2 commented Oct 18, 2024

cmsbuild commented Oct 18, 2024

mandrenguyen commented Oct 18, 2024

	# Alpaka
	from Configuration.ProcessModifiers.alpaka_cff import alpaka
	alpaka.toReplaceWith(_HgcalLocalRecoUnseededSequence,
	cms.Sequence(
	hltHgcalDigis
	+ hltHGCalUncalibRecHit
	+ hltHGCalRecHit+hltParticleFlowRecHitHGC
	+ hltHgcalSoARecHitsProducer
	+ hltHgcalSoARecHitsLayerClustersProducer
	+ hltHgcalSoALayerClustersProducer
	+ hltHgCalLayerClustersFromSoAProducer
	+ hltHgcalLayerClustersHSci
	+ hltHgcalLayerClustersHSi
	+ hltHgcalMergeLayerClusters
	)
	)

	alpaka.toModify(hltHgcalMergeLayerClusters,
	layerClustersEE = cms.InputTag("hltHgCalLayerClustersFromSoAProducer"),
	time_layerclustersEE = cms.InputTag("hltHgCalLayerClustersFromSoAProducer", "timeLayerCluster"))

	(alpaka & ticl_superclustering_mustache_ticl).toModify(hltHgcalMergeLayerClusters,
	layerClustersEE = cms.InputTag("hltHgCalLayerClustersFromSoAProducer"),
	time_layerclustersEE = cms.InputTag("hltHgCalLayerClustersFromSoAProducer", "timeLayerCluster"))

Enable DNN supercluster in Phase2 HLT #46010

Enable DNN supercluster in Phase2 HLT #46010

Conversation

RSalvatico commented Sep 17, 2024

PR description:

PR validation:

cmsbuild commented Sep 17, 2024 • edited Loading

cmsbuild commented Sep 17, 2024

cmsbuild commented Sep 17, 2024

felicepantaleo commented Sep 17, 2024

felicepantaleo commented Sep 17, 2024

felicepantaleo commented Sep 17, 2024

cmsbuild commented Sep 17, 2024

Comparison Summary

mmusich commented Sep 17, 2024

mmusich left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

RSalvatico Sep 17, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mmusich commented Oct 9, 2024

mmusich commented Oct 9, 2024

cmsbuild commented Oct 9, 2024

Comparison Summary

mmusich commented Oct 17, 2024

jfernan2 commented Oct 17, 2024

Moanwar commented Oct 17, 2024

AdrianoDee commented Oct 17, 2024

RSalvatico commented Oct 18, 2024

jfernan2 commented Oct 18, 2024

cmsbuild commented Oct 18, 2024

mandrenguyen commented Oct 18, 2024

cmsbuild commented Sep 17, 2024 •

edited

Loading

RSalvatico Sep 17, 2024 •

edited

Loading