[L1T] Update L1T MET and Add JUMP algorithm (Phase 2) #48308

JunwonTomOh · 2025-06-13T06:33:37Z

PR description:
This PR introduces two new L1 MET calculation algorithms and related implementations:

A new MET emulator using polynomial interpolation has been added. This algorithm was not present in the existing emulator framework and enables smoother, resource-efficient computation of MET using a fitted interpolation curve.

A corresponding extension was made to the existing MET producer to include this new interpolation-based method. This allows users to evaluate and deploy the interpolated MET algorithm directly in CMSSW workflows.

The PR also adds a new algorithm named JUMP (Jet Uncertainty-aware MET Prediction), which uses jet energy resolution information to correct MET in high pileup conditions. Both the producer and the emulator for JUMP were developed from scratch.
DPS Note publicly available on CDS: CMS DP-2025/023

These additions aim to enhance MET performance in challenging pileup scenarios and facilitate hardware emulation and validation of new MET strategies.

PR validation:
All new algorithms have been tested using the FastPUPPI
Used /eos/cms/store/cmst3/group/l1tr/FastPUPPI/14_2_X/fpinputs_140X/v0/TTToSemileptonic_PU200

To support the validation of these algorithms, reference plots prepared by @thesps can be found here:
https://ssummers.web.cern.ch/CMS/TauJetsMET/V44JUMP/ba40f8a/

@thesps

cmsbuild · 2025-06-13T06:33:59Z

cms-bot internal usage

cmsbuild · 2025-06-13T06:36:24Z

-code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-48308/45177

Code check has found code style and quality issues which could be resolved by applying following patch(s)

code-checks:
https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-48308/45177/code-checks.patch
e.g. curl -k https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-48308/45177/code-checks.patch | patch -p1
You can also run scram build code-checks to apply code checks directly
code-format:
https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-48308/45177/code-format.patch
e.g. curl -k https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-48308/45177/code-format.patch | patch -p1
You can also run scram build code-format to apply code format directly

cmsbuild · 2025-06-13T07:05:28Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-48308/45179

cmsbuild · 2025-06-13T07:05:45Z

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

It involves the following packages:

L1Trigger/Phase2L1ParticleFlow (l1, upgrade)

@BenjaminRS, @Moanwar, @cmsbuild, @quinnanm, @srimanob, @subirsarkar can you please review it and eventually sign? Thanks.
@Martin-Grunewald, @missirol, @mmusich 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

BenjaminRS · 2025-06-16T13:38:40Z

test parameters:

workflow = 29634.78
relvals_opt = --what upgrade

BenjaminRS · 2025-06-16T13:39:08Z

please test

BenjaminRS · 2025-06-16T14:32:31Z

@JunwonTomOh can you update the title of this PR please, following the L1T recommendations, to:
[L1T] Update L1T MET and Add JUMP algorithm (Phase 2)

JunwonTomOh · 2025-06-16T15:49:59Z

@BenjaminRS Thank you for the comments. I've just changed.

cmsbuild · 2025-06-16T15:59:51Z

+1

Size: This PR adds an extra 16KB to repository
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-ab62aa/46762/summary.html
COMMIT: 070951d
CMSSW: CMSSW_15_1_X_2025-06-16-1100/el8_amd64_gcc12
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/48308/46762/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

You potentially removed 5 lines from the logs
Reco comparison results: 12 differences found in the comparisons
DQMHistoTests: Total files compared: 51
DQMHistoTests: Total histograms compared: 4181335
DQMHistoTests: Total failures: 58
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 4181257
DQMHistoTests: Total skipped: 20
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 50 files compared)
Checked 220 log files, 188 edm output root files, 51 DQM output files
TriggerResults: no differences found

quinnanm · 2025-06-23T22:38:18Z

L1Trigger/Phase2L1ParticleFlow/interface/jetmet/L1PFJUMPEmulator.h

+
+    L1METEmu::eta_t eta_edges[4];
+    float eta_boundaries[4] = {1.3, 1.7, 2.5, 3.0};
+    for (uint i = 0; i < 4; i++) {


should uint be unsigned int? also other places in the code

In this case, we can use the int type, but when we compare with .size(), it seems like we should use the uint type. For example, at line 76 in L1Trigger/Phase2L1ParticleFlow/interface/jetmet/L1PFJUMPEmulator.h, uint is used like this.

for (uint i = 0; i < jets.size(); i++) { Get_dPt(jets[i], each_dPx2, each_dPy2); sum_dPx2 += each_dPx2; sum_dPy2 += each_dPy2; }

I'll change the uint into int type if possible. Thanks for the comment and please feel free to share any advise.

ok if you think it is more appropriate feel free to use uint. this was more of a question than a strict suggestion!

quinnanm · 2025-06-23T22:39:10Z

L1Trigger/Phase2L1ParticleFlow/interface/jetmet/L1PFJUMPEmulator.h

+                                       : inMet_xy.hwPy - L1METEmu::proj2_t(sqrt(dPy_2.to_float()));
+  L1METEmu::pxpy_to_ptphi(outMet_xy, outMet);
+
+  return;


not a big deal, but dont need return; for a void function

quinnanm · 2025-06-23T22:40:09Z

L1Trigger/Phase2L1ParticleFlow/interface/jetmet/L1PFMetEmulator.h

+    poly_t cos2_par0[16] = {-1.00007,
+                            -0.924181,
+                            -0.707596,
+                            -0.382902,
+                            -0.000618262,
+                            0.382137,
+                            0.707056,
+                            0.923708,
+                            1.00007,
+                            0.924181,
+                            0.707594,
+                            0.383285,
+                            0.000188727,
+                            -0.382139,
+                            -0.706719,
+                            -0.923708};
+    poly_t cos2_par1[16] = {9.164680268990924e-06,
+                            0.0017064607695524156,
+                            0.0031441321076514446,
+                            0.004079929656016374,
+                            0.004437063290882583,
+                            0.004095969231842202,
+                            0.0031107221424451436,
+                            0.001689531075808071,
+                            -9.161756842493832e-06,
+                            -0.001706456406229286,
+                            -0.003143961938049376,
+                            -0.004103015998697129,
+                            -0.004411145151490469,
+                            -0.0040958165155326525,
+                            -0.0031310072316764474,
+                            -0.001689531075808071};
+    poly_t cos2_par2[16] = {9.319674765430664e-06,
+                            7.871694899063284e-06,
+                            5.222989318251642e-06,
+                            2.0256106486379287e-06,
+                            -1.9299417402361656e-06,
+                            -5.35167113952279e-06,
+                            -7.740062096537953e-06,
+                            -9.348822844786505e-06,
+                            -9.319674765430664e-06,
+                            -7.871694899063284e-06,
+                            -5.225331064666252e-06,
+                            -1.780776301343235e-06,
+                            1.6556927733433181e-06,
+                            5.3495197789955455e-06,
+                            7.954684107366423e-06,
+                            9.348822844786505e-06};
+
+    poly_t sin2_par0[16] = {0.000524872,
+                            -0.382229,
+                            -0.706791,
+                            -0.923959,
+                            -1.00008,
+                            -0.924156,
+                            -0.707264,
+                            -0.383199,
+                            -0.000525527,
+                            0.382228,
+                            0.706792,
+                            0.923752,
+                            1.00013,
+                            0.924155,
+                            0.707535,
+                            0.3832};
+    poly_t sin2_par1[16] = {-0.004431478237276202,
+                            -0.00409041472149773,
+                            -0.0031267268116859314,
+                            -0.00167440343451641,
+                            9.741773386162849e-06,
+                            0.0017049641497188307,
+                            0.00312406082125351,
+                            0.0040978672774037465,
+                            0.004431478237276202,
+                            0.00409041472149773,
+                            0.0031266351819002015,
+                            0.0016868781753450394,
+                            -1.249302315254411e-05,
+                            -0.001704846339994321,
+                            -0.003140405829698437,
+                            -0.0040978672774037465};
+    poly_t sin2_par2[16] = {1.870674613498914e-06,
+                            5.292404012785538e-06,
+                            7.909829192302831e-06,
+                            9.188746390688592e-06,
+                            9.313525301268721e-06,
+                            7.887020962996302e-06,
+                            5.435897856093815e-06,
+                            1.8358587462761668e-06,
+                            -1.870668901922293e-06,
+                            -5.292404012785538e-06,
+                            -7.908420336736317e-06,
+                            -9.320836119343602e-06,
+                            -9.284396260501616e-06,
+                            -7.88869635880513e-06,
+                            -5.262894200243701e-06,
+                            -1.835864457852788e-06};


Would it be cleaner to have these numbers in an external json file in cms-data?

quinnanm · 2025-06-23T22:40:33Z

L1Trigger/Phase2L1ParticleFlow/interface/jetmet/L1PFMetEmulator.h

+    hls_met.hwPhi = phi_t(ap_fixed<26, 11>(l1ct::Scales::makeGlbPhi(hls::atan2(met_xy.hwPy, met_xy.hwPx))));
+#endif
+
+    return;


unused return;

quinnanm · 2025-06-23T22:40:52Z

L1Trigger/Phase2L1ParticleFlow/interface/jetmet/L1PFMetEmulator.h

+  L1METEmu::Sum_Particles(particles_xy, met_xy);
+  L1METEmu::pxpy_to_ptphi(met_xy, out_met);
+
+  return;


unused return;

quinnanm · 2025-06-23T22:41:40Z

L1Trigger/Phase2L1ParticleFlow/interface/jetmet/L1PFMetEmulator.h

+    return;
+  }
+
+  inline void met_format(l1ct::Sum d, ap_uint<l1gt::Sum::BITWIDTH>& q) {


function inputs should be const refs where possible

quinnanm · 2025-06-23T22:42:13Z

L1Trigger/Phase2L1ParticleFlow/plugins/L1JUMPProducer.cc

+
+#include "L1Trigger/Phase2L1ParticleFlow/interface/jetmet/L1PFJUMPEmulator.h"
+
+using namespace l1t;


please avoid using namespace l1t in favor of l1t::func if possible

quinnanm · 2025-06-23T22:42:38Z

L1Trigger/Phase2L1ParticleFlow/plugins/L1JUMPProducer.cc

+  iEvent.put(std::move(JUMPCollection));
+}
+
+void L1JUMPProducer::CalcJUMP_HLS(l1t::EtSum& metVector,


function inputs should be const refs where possible

quinnanm · 2025-06-23T22:42:57Z

L1Trigger/Phase2L1ParticleFlow/plugins/L1JUMPProducer.cc

+  outMet_Vector.SetEta(0);
+}
+
+std::vector<l1ct::Jet> L1JUMPProducer::convertEDMToHW(std::vector<l1t::PFJet> edmJets) const {


function inputs should be const refs where possible

quinnanm · 2025-06-23T22:44:11Z

L1Trigger/Phase2L1ParticleFlow/plugins/L1JUMPProducer.cc

+
+std::vector<l1ct::Jet> L1JUMPProducer::convertEDMToHW(std::vector<l1t::PFJet> edmJets) const {
+  std::vector<l1ct::Jet> hwJets;
+  std::for_each(edmJets.begin(), edmJets.end(), [&](l1t::PFJet jet) {


better as std::for_each(edmJets.begin(), edmJets.end(), [&](const l1t::PFJet& jet) { but not a big deal

BenjaminRS · 2025-09-09T20:06:07Z

please test

cmsbuild · 2025-09-09T22:40:25Z

+1

Size: This PR adds an extra 32KB to repository
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-ab62aa/48039/summary.html
COMMIT: af82bf2
CMSSW: CMSSW_15_1_X_2025-09-09-1100/el8_amd64_gcc12
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/48308/48039/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:

@tihsu99 Include scouting EGM in Online DQM stream #48850

You can see more details here:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-ab62aa/48039/git-recent-commits.json
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-ab62aa/48039/git-merge-result

Comparison Summary

Summary:

You potentially added 14 lines to the logs
Reco comparison results: 0 differences found in the comparisons
DQMHistoTests: Total files compared: 51
DQMHistoTests: Total histograms compared: 4248921
DQMHistoTests: Total failures: 8
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 4248893
DQMHistoTests: Total skipped: 20
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 50 files compared)
Checked 220 log files, 188 edm output root files, 51 DQM output files
TriggerResults: no differences found

cmsbuild · 2025-09-10T06:56:05Z

Milestone for this pull request has been moved to CMSSW_16_0_X. Please open a backport if it should also go in to CMSSW_15_1_X.

BenjaminRS

Thanks for the modifications. Looks good.

BenjaminRS · 2025-09-10T13:25:43Z

+l1

Moanwar · 2025-09-10T13:27:13Z

+Upgrade

cmsbuild · 2025-09-10T13:27:35Z

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. @ftenchini, @mandrenguyen, @sextonkennedy (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/L1Trigger-Phase2L1ParticleFlow#9

mandrenguyen · 2025-09-10T19:26:43Z

+1

mandrenguyen · 2025-09-11T06:42:17Z

This PR, and the corresponding cms-data one, are causing 16_0_X IBs to fail:
cms-data/L1Trigger-Phase2L1ParticleFlow#9 (comment)
Please have a look.

Update L1T MET and Add JUMP algorithm

db7e90b

cmsbuild added this to the CMSSW_15_1_X milestone Jun 13, 2025

cmsbuild added l1-pending pending-signatures tests-pending orp-pending upgrade-pending code-checks-pending labels Jun 13, 2025

cmsbuild added code-checks-rejected and removed code-checks-pending labels Jun 13, 2025

Apply code checks and format

070951d

cmsbuild added code-checks-pending and removed code-checks-rejected labels Jun 13, 2025

cmsbuild added code-checks-approved and removed code-checks-pending labels Jun 13, 2025

cmsbuild added tests-started and removed tests-pending labels Jun 16, 2025

JunwonTomOh changed the title ~~Update L1T MET and Add JUMP algorithm~~ [L1T] Update L1T MET and Add JUMP algorithm (Phase 2) Jun 16, 2025

cmsbuild added tests-approved and removed tests-started labels Jun 16, 2025

quinnanm reviewed Jun 23, 2025

View reviewed changes

cmsbuild added tests-started and removed tests-pending labels Sep 9, 2025

cmsbuild added tests-approved and removed tests-started labels Sep 9, 2025

cmsbuild modified the milestones: CMSSW_15_1_X, CMSSW_16_0_X Sep 10, 2025

BenjaminRS approved these changes Sep 10, 2025

View reviewed changes

cmsbuild added l1-approved and removed l1-pending labels Sep 10, 2025

cmsbuild added fully-signed upgrade-approved and removed pending-signatures upgrade-pending labels Sep 10, 2025

cmsbuild added orp-approved and removed orp-pending labels Sep 10, 2025

cmsbuild merged commit b264073 into cms-sw:master Sep 10, 2025
11 checks passed

mandrenguyen mentioned this pull request Sep 11, 2025

16_0_X IB failures related to missing cms-data file in L1Trigger-Phase2L1ParticleFlow #48897

Closed

JunwonTomOh mentioned this pull request Nov 18, 2025

Synchronize L1 MET code with CMSSW main branch, fix phi calculation for HLS testbench, and temporarily add JSON files p2l1pfp/cmssw#143

Open


		#include "L1Trigger/Phase2L1ParticleFlow/interface/jetmet/L1PFJUMPEmulator.h"

		using namespace l1t;

[L1T] Update L1T MET and Add JUMP algorithm (Phase 2) #48308

[L1T] Update L1T MET and Add JUMP algorithm (Phase 2) #48308

Uh oh!

Conversation

JunwonTomOh commented Jun 13, 2025

Uh oh!

cmsbuild commented Jun 13, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

cmsbuild commented Jun 13, 2025

Uh oh!

cmsbuild commented Jun 13, 2025

Uh oh!

cmsbuild commented Jun 13, 2025

Uh oh!

BenjaminRS commented Jun 16, 2025

Uh oh!

BenjaminRS commented Jun 16, 2025

Uh oh!

BenjaminRS commented Jun 16, 2025

Uh oh!

JunwonTomOh commented Jun 16, 2025

Uh oh!

cmsbuild commented Jun 16, 2025

Comparison Summary

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

BenjaminRS commented Sep 9, 2025

Uh oh!

cmsbuild commented Sep 9, 2025

Comparison Summary

Uh oh!

cmsbuild commented Sep 10, 2025

Uh oh!

BenjaminRS left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

BenjaminRS commented Sep 10, 2025

Uh oh!

Moanwar commented Sep 10, 2025

Uh oh!

cmsbuild commented Sep 10, 2025

Uh oh!

mandrenguyen commented Sep 10, 2025

Uh oh!

Uh oh!

mandrenguyen commented Sep 11, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

cmsbuild commented Jun 13, 2025 •

edited

Loading

BenjaminRS left a comment •

edited

Loading