cms-sw
modified for Phase2 DeepJet retraining model
PR description:
The BTV@HLT group has a Phase2 DeepJet retrained model. The target for this retraining is to retrain DeepJet and recreate/Improve TDR performance. The retraining online performance is compared to the TDR/offline model. Please see the details here and here. This PR is associated with the Phase2 DeepJet retraining model integration PR.
PR validation:
The model was tested locally in CMSSW_13_1_0.
Follow the Phase 2 HLT simplified menu and edmConfigDump to get the configuration file and the HLT dump phase2_hlt.py, and in process.hltPfDeepFlavourJetTagsModEta2p4 and
process.hltPfDeepFlavourJetTags, change the model_path to retained model.
After running cmsRun phase2_hlt.py, the same warnings from InclusiveCandidateVertexFinder and EcalRecHitProducer are observed as running with RecoBTag/Combined/data/DeepFlavourV01_PhaseII/model.onnx, which is the default model.
Please see the details in the presentation.
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:
It'll have to be backported to CMSSW_14_0_X.
Before submitting your pull requests, make sure you followed this checklist:
- verify that the PR is really intended for the chosen branch
- verify that changes follow CMS Naming, Coding, And Style Rules
- verify that the PR passes the basic test procedure suggested in the CMSSW PR instructions
+code-checks
Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-44431/39505
- This PR adds an extra 28KB to repository
A new Pull Request was created by @hsinweihsia for master.
It involves the following packages:
- HLTrigger/Configuration (hlt)
@Martin-Grunewald, @cmsbuild, @mmusich can you please review it and eventually sign? Thanks. @rovere, @Martin-Grunewald, @SohamBhattacharya, @missirol, @silviodonato this is something you requested to watch as well. @antoniovilela, @rappoccio, @sextonkennedy you are the release manager for this.
cms-bot commands are listed here
+code-checks
Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-44431/39506
- This PR adds an extra 32KB to repository
Pull request #44431 was updated. @mmusich, @Martin-Grunewald, @cmsbuild can you please check and sign again.
test parameters:
- pull_request = https://github.com/cms-data/RecoBTag-Combined/pull/56
The BTV@HLT group has a Phase2 DeepJet retrained model.
@hsinweihsia please edit the PR description to link the study that motivated this update. Please add also a description of the validation done and plans for a backport. Thank you
+1
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-7e7784/38186/summary.html
COMMIT: e80e4d8be80049aadb8073170611fab69f6a5f73
CMSSW: CMSSW_14_1_X_2024-03-15-2300/el8_amd64_gcc12
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/44431/38186/install.sh to create a dev area with all the needed externals and cmssw changes.
Comparison Summary
Summary:
- You potentially added 106 lines to the logs
- Reco comparison results: 39 differences found in the comparisons
- DQMHistoTests: Total files compared: 48
- DQMHistoTests: Total histograms compared: 3297369
- DQMHistoTests: Total failures: 0
- DQMHistoTests: Total nulls: 0
- DQMHistoTests: Total successes: 3297349
- DQMHistoTests: Total skipped: 20
- DQMHistoTests: Total Missing objects: 0
- DQMHistoSizes: Histogram memory added: 0.0 KiB( 47 files compared)
- Checked 202 log files, 165 edm output root files, 48 DQM output files
- TriggerResults: no differences found
I have no reason to suspect issues with this PR (test results are technically fine), but unfortunately we don't seem to have any DQM / validation for the phase-2 b-tagging objects in the DQM matrix (see also issue https://github.com/cms-sw/cmssw/issues/39362). @hsinweihsia is is something that BTV POG can work on to improve our monitoring capabilities ? For sign-off I'll wait explicit green-light from TSG upgrade @rovere @SohamBhattacharya
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, @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#56
@hsinweihsia @rovere @SohamBhattacharya
It'll have to be backported to CMSSW_14_0_X.
do you still foresee this to happen?
Re-ping @hsinweihsia @rovere @SohamBhattacharya Do you still expect this in 14_0? Should I wait for it, or start relvals as soon as I have a new release. Thx.
