Add the Scouting Glo-ParT inference facility #47801

yiyangzha · 2025-04-07T19:52:18Z

PR description:

This PR adds the Scouting Glo-ParT model's inference facility into CMSSW, prepared for Scouting NanoAOD official production.
Model performance details are provided in these [slides] (accessible via CMS). This would add 13 taggers and 4 mass regression correctors to Scouting NanoAOD.

Scouting Global Particle Transformer (Glo-ParT) is an inclusive tagging model for AK8 scouting PFjets. It functions as both a global tagger and a mass regression model for AK8 scouting PFjets and can also be utilized as a pre-trained model. Further details can be found in the slides.

Please test this PR with cms-data/RecoBTag-Combined#67.

PR validation:

The PR passed the tests listed at https://cms-sw.github.io/PRWorkflow.html.

If this PR is a backport please specify the original PR and why you need to backport that PR. If this PR will be backported please specify to which release cycle the backport is meant for:

Will be backported to CMSSW_15_0_X: #47853

cmsbuild · 2025-04-07T19:52:46Z

cms-bot internal usage

cmsbuild · 2025-04-07T19:54:31Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-47801/44397

There are other open Pull requests which might conflict with changes you have proposed:
- File RecoBTag/FeatureTools/plugins/DeepBoostedJetTagInfoProducer.cc modified in PR(s): update PackedCandidate::dzError to return what it should in track parameterization #45612

cmsbuild · 2025-04-07T19:54:55Z

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

It involves the following packages:

PhysicsTools/NanoAOD (xpog)
RecoBTag/FeatureTools (reconstruction)

@cmsbuild, @ftorrresd, @hqucms, @jfernan2, @mandrenguyen can you please review it and eventually sign? Thanks.
@AlexDeMoor, @Ming-Yan, @Senphy, @andrzejnovak, @castaned, @gpetruc, @hqucms, @missirol 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

jfernan2 · 2025-04-08T11:07:22Z

enable nano

jfernan2 · 2025-04-08T11:07:43Z

please test

yiyangzha · 2025-04-08T11:58:07Z

I think it's caused by not testing with cms-data/RecoBTag-Combined#67
Could you help test again including this test parameter? Thanks!

cmsbuild · 2025-04-08T13:30:30Z

-1

Failed Tests: RelVals-INPUT RelVals-NANO
Size: This PR adds an extra 40KB to repository
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-0cee3f/45439/summary.html
COMMIT: 44eb104
CMSSW: CMSSW_15_1_X_2025-04-07-2300/el8_amd64_gcc12
Additional Tests: NANO
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/47801/45439/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

You potentially added 1 lines to the logs
Reco comparison results: 8 differences found in the comparisons
DQMHistoTests: Total files compared: 50
DQMHistoTests: Total histograms compared: 3913985
DQMHistoTests: Total failures: 11
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 3913954
DQMHistoTests: Total skipped: 20
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 49 files compared)
Checked 215 log files, 184 edm output root files, 50 DQM output files
TriggerResults: no differences found

yiyangzha · 2025-04-08T13:38:21Z

I think it's caused by not testing with cms-data/RecoBTag-Combined#67 Could you help test again including this test parameter? Thanks!

I think this should be the reasons for Failed Tests.

ftorrresd · 2025-04-08T19:04:52Z

test parameters:
pull_request = cms-data/RecoBTag-Combined#67
enable = nano

ftorrresd · 2025-04-08T19:09:10Z

please test

cmsbuild · 2025-04-08T22:01:40Z

+1

Size: This PR adds an extra 16KB to repository
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-0cee3f/45455/summary.html
COMMIT: 44eb104
CMSSW: CMSSW_15_1_X_2025-04-08-1100/el8_amd64_gcc12
Additional Tests: NANO
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/47801/45455/install.sh to create a dev area with all the needed externals and cmssw changes.

The following merge commits were also included on top of IB + this PR after doing git cms-merge-topic:

You can see more details here:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-0cee3f/45455/git-recent-commits.json
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-0cee3f/45455/git-merge-result

Comparison Summary

Summary:

You potentially added 4 lines to the logs
ROOTFileChecks: Some differences in event products or their sizes found
Reco comparison results: 62 differences found in the comparisons
DQMHistoTests: Total files compared: 50
DQMHistoTests: Total histograms compared: 3912623
DQMHistoTests: Total failures: 1431
DQMHistoTests: Total nulls: 219
DQMHistoTests: Total successes: 3910953
DQMHistoTests: Total skipped: 20
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 575.308 KiB( 49 files compared)
DQMHistoSizes: changed ( 145.014,... ): 26.944 KiB HLT/Filters
DQMHistoSizes: changed ( 145.014,... ): 16.803 KiB HLT/TAU
DQMHistoSizes: changed ( 16834.0,... ): 49.804 KiB HLT/TAU
DQMHistoSizes: changed ( 16834.0,... ): 36.082 KiB HLT/HLTEgammaValidation
DQMHistoSizes: changed ( 16834.0,... ): -0.164 KiB L1T/L1TStage2uGT
Checked 215 log files, 184 edm output root files, 50 DQM output files
TriggerResults: found differences in 10 / 48 workflows

NANO Comparison Summary

Summary:

You potentially removed 461 lines from the logs
ROOTFileChecks: Some differences in event products or their sizes found
Reco comparison results: 0 differences found in the comparisons
DQMHistoTests: Total files compared: 23
DQMHistoTests: Total histograms compared: 90509
DQMHistoTests: Total failures: 0
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 90509
DQMHistoTests: Total skipped: 0
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 22 files compared)
Checked 113 log files, 65 edm output root files, 23 DQM output files
TriggerResults: no differences found

Nano size comparison Summary:

Sample	kb/ev	ref kb/ev	diff kb/ev	ev/s/thd	ref ev/s/thd	diff rate	mem/thd	ref mem/thd
2500.001	3.115	3.115	0.000 ( +0.0% )	6.96	6.77	+2.8%	2.590	2.581
2500.002	3.231	3.231	0.000 ( +0.0% )	6.19	6.07	+1.9%	3.024	3.017
2500.003	3.172	3.172	0.000 ( +0.0% )	6.45	6.40	+0.8%	2.998	2.997
2500.011	1.647	1.647	0.000 ( +0.0% )	11.67	11.55	+1.0%	2.672	2.663
2500.012	2.185	2.185	0.000 ( +0.0% )	6.50	6.40	+1.6%	2.853	2.843
2500.013	2.002	2.002	0.000 ( +0.0% )	9.15	9.08	+0.8%	2.763	2.755
2500.021	0.022	0.022	0.000 ( +0.0% )	1.86	1.90	-2.0%	2.709	2.708
2500.022	0.022	0.022	0.000 ( +0.0% )	1.81	1.82	-0.5%	2.697	2.706
2500.023	0.022	0.022	0.000 ( +0.0% )	1.75	1.80	-2.7%	2.568	2.570
2500.024	0.022	0.022	0.000 ( +0.0% )	1.47	1.50	-2.2%	2.810	2.806
2500.031	0.035	0.035	0.000 ( +0.0% )	1.62	1.67	-3.2%	2.763	2.765
2500.032	0.036	0.036	0.000 ( +0.0% )	1.63	1.66	-1.5%	2.713	2.735
2500.033	0.037	0.037	0.000 ( +0.1% )	1.55	1.61	-3.7%	2.797	2.813
2500.034	0.036	0.036	0.000 ( +0.0% )	1.59	1.60	-0.6%	2.793	2.787
2500.101	2.872	2.872	0.000 ( +0.0% )	16.05	15.96	+0.6%	2.649	2.647
2500.111	1.474	1.474	0.000 ( +0.0% )	31.40	31.03	+1.2%	2.344	2.338
2500.112	1.896	1.896	0.000 ( +0.0% )	26.89	24.95	+7.8%	2.413	2.406
2500.131	0.758	0.750	0.007 ( +1.0% )	33.96	36.89	-7.9%	1.634	1.500
2500.201	2.706	2.706	0.000 ( +0.0% )	12.53	13.20	-5.1%	2.219	2.216
2500.211	1.845	1.845	0.000 ( +0.0% )	26.83	26.65	+0.7%	2.411	2.417
2500.212	2.243	2.243	0.000 ( +0.0% )	23.16	22.98	+0.8%	2.500	2.492
2500.221	2.141	2.141	0.000 ( +0.0% )	14.12	13.94	+1.3%	2.136	2.129
2500.222	3.516	3.516	0.000 ( +0.0% )	13.40	13.25	+1.1%	2.230	2.221
2500.223	10.328	10.328	0.000 ( +0.0% )	4.63	4.67	-0.8%	2.399	2.395
2500.224	6.622	6.622	0.000 ( +0.0% )	1.26	1.25	+0.8%	2.348	2.341
2500.225	6.671	6.671	0.000 ( +0.0% )	1.18	1.17	+0.9%	2.572	2.553
2500.226	3.210	3.210	0.000 ( +0.0% )	13.60	13.47	+0.9%	2.216	2.223
2500.227	1.463	1.442	0.021 ( +1.5% )	19.32	23.53	-17.9%	1.579	1.444
2500.228	4.075	4.045	0.030 ( +0.7% )	8.17	8.88	-8.0%	2.466	2.321
2500.231	1.516	1.516	0.000 ( +0.0% )	21.53	22.31	-3.5%	2.326	2.311
2500.232	2.502	2.502	0.000 ( +0.0% )	21.11	21.89	-3.6%	2.409	2.396
2500.233	5.422	5.422	0.000 ( +0.0% )	7.07	7.04	+0.4%	2.579	2.571
2500.234	3.928	3.928	0.000 ( +0.0% )	1.62	1.59	+1.9%	2.494	2.477
2500.235	3.960	3.960	0.000 ( +0.0% )	1.54	1.51	+2.0%	2.678	2.690
2500.236	2.292	2.292	0.000 ( +0.0% )	22.42	22.19	+1.0%	2.407	2.394
2500.237	1.030	1.018	0.012 ( +1.1% )	30.28	34.40	-12.0%	1.582	1.454
2500.238	2.501	2.477	0.024 ( +1.0% )	14.99	16.97	-11.7%	2.622	2.479
2500.241	9.404	9.404	0.000 ( +0.0% )	7.85	7.78	+0.9%	1.932	1.927
2500.242	10.331	10.331	0.000 ( +0.0% )	1.69	1.67	+0.7%	1.733	1.729
2500.243	2.712	2.712	0.000 ( +0.0% )	15.99	16.04	-0.3%	1.064	1.065
2500.244	486.016	486.016	0.000 ( +0.0% )	1.16	1.15	+0.9%	1.716	1.697
2500.245	826.413	826.413	0.000 ( +0.0% )	1.55	1.53	+0.9%	1.688	1.685
2500.251	645.333	645.333	0.000 ( +0.0% )	1.66	1.65	+1.0%	1.796	1.786
2500.301	0.021	0.021	0.000 ( +0.0% )	1.79	1.71	+4.5%	2.821	2.824
2500.311	0.036	0.036	0.000 ( +0.0% )	1.72	1.70	+1.2%	2.770	2.769
2500.901	1.819	1.819	0.000 ( +0.0% )	47.59	45.09	+5.5%	1.440	1.439
2500.902	1.665	1.665	0.000 ( +0.0% )	49.21	46.32	+6.2%	1.344	1.344
2500.911	14.345	14.345	0.000 ( +0.0% )	8.55	8.09	+5.7%	1.094	1.092
2500.912	0.171	0.240	-0.069 ( -28.7% )	4.00	3.18	+25.6%	0.852	0.848
2500.913	0.110	0.110	0.000 ( +0.0% )	2.64	2.62	+0.7%	0.852	0.851

patinkaew · 2025-04-17T10:26:49Z

Hi all,

FYI: ScoutingNano related workflows are: 2500.131,2500.227,2500.228,2500.237,2500.238.
2500.228, 2500.238 are @Prompt+@Scout flavour, explaining the large event size.
Nonetheless, from the tests, there is no significant increase in event size for all five workflows above.

ftorrresd · 2025-04-19T12:37:40Z

+1

jfernan2 · 2025-04-19T13:14:28Z

+1

cmsbuild · 2025-04-19T13:14:51Z

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. @sextonkennedy, @antoniovilela, @mandrenguyen, @rappoccio (and backports should be raised in the release meeting by the corresponding L2)
Notice This PR was tested with additional Pull Request(s), please also merge them if necessary: cms-data/RecoBTag-Combined#67

mandrenguyen · 2025-04-21T07:38:38Z

+1

[15_0_X] Backport of #47801 (Add the Scouting Glo-ParT inference facility)

Add the Scouting Glo-ParT inference facility

44eb104

cmsbuild added this to the CMSSW_15_1_X milestone Apr 7, 2025

cmsbuild added reconstruction-pending pending-signatures tests-pending orp-pending code-checks-pending xpog-pending labels Apr 7, 2025

cmsbuild added code-checks-approved and removed code-checks-pending labels Apr 7, 2025

cmsbuild added tests-started and removed tests-pending labels Apr 8, 2025

cmsbuild added tests-rejected and removed tests-started labels Apr 8, 2025

cmsbuild added the requires-external label Apr 8, 2025

cmsbuild added tests-started and removed tests-rejected labels Apr 8, 2025

cmsbuild added tests-approved and removed tests-started labels Apr 8, 2025

yiyangzha mentioned this pull request Apr 12, 2025

[15_0_X] Backport of #47801 (Add the Scouting Glo-ParT inference facility) #47853

Merged

cmsbuild added xpog-approved and removed xpog-pending labels Apr 19, 2025

cmsbuild added reconstruction-approved fully-signed and removed reconstruction-pending pending-signatures labels Apr 19, 2025

cmsbuild mentioned this pull request Apr 19, 2025

Add Scouting Glo-ParT ONNX model for Run-3 PFScouting cms-data/RecoBTag-Combined#67

Merged

cmsbuild added orp-approved and removed orp-pending labels Apr 21, 2025

cmsbuild merged commit c3122c5 into cms-sw:master Apr 21, 2025
14 checks passed

patinkaew mentioned this pull request Apr 23, 2025

ScoutingNano Event Content Changes for 2022-25 #47934

Merged

cmsbuild added a commit that referenced this pull request Apr 23, 2025

Merge pull request #47853 from yiyangzha/CMSSW_15_0_X

fecb586

[15_0_X] Backport of #47801 (Add the Scouting Glo-ParT inference facility)

Add the Scouting Glo-ParT inference facility #47801

Add the Scouting Glo-ParT inference facility #47801

Conversation

yiyangzha commented Apr 7, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR description:

PR validation:

If this PR is a backport please specify the original PR and why you need to backport that PR. If this PR will be backported please specify to which release cycle the backport is meant for:

Uh oh!

cmsbuild commented Apr 7, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

cmsbuild commented Apr 7, 2025

Uh oh!

cmsbuild commented Apr 7, 2025

Uh oh!

jfernan2 commented Apr 8, 2025

Uh oh!

jfernan2 commented Apr 8, 2025

Uh oh!

yiyangzha commented Apr 8, 2025

Uh oh!

cmsbuild commented Apr 8, 2025

Comparison Summary

Uh oh!

yiyangzha commented Apr 8, 2025

Uh oh!

ftorrresd commented Apr 8, 2025

Uh oh!

ftorrresd commented Apr 8, 2025

Uh oh!

cmsbuild commented Apr 8, 2025

Comparison Summary

NANO Comparison Summary

Uh oh!

patinkaew commented Apr 17, 2025

Uh oh!

ftorrresd commented Apr 19, 2025

Uh oh!

jfernan2 commented Apr 19, 2025

Uh oh!

cmsbuild commented Apr 19, 2025

Uh oh!

mandrenguyen commented Apr 21, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

yiyangzha commented Apr 7, 2025 •

edited

Loading

cmsbuild commented Apr 7, 2025 •

edited

Loading