Backport: Fixes to L2Tau CNN and deepTauID for HLT

[azotz]

PR description:

This PR fixes critical issues found in L2NNTag modules and deepTauID for HLT. The PR contains the following changes:

• Add default values for configurable parameters where those have not been set to allow correct parsing to ConfDB;
• Allow a pass-through at L2 for all L1 taus with pt above a configurable threshold in L2NNTauTag to avoid observed drop of efficiency for taus with gen_vis_pt > 500 GeV (for gen_vis_pt < 500 GeV the algo efficiency is above 90%). We estimated that the rate with the pass-though threshold of 250 GeV (default) does not significantly change: for the 3 kHz Working Point (WP) it goes from 2999.7 Hz to 3026.4 Hz (0.9%); changes are similar for other WPs.

The changes are simple fixes and do not change behavior of the modules, but on the other hand are critical as without them L2Tau CNN and deepTauID cannot be deployed at HLT.

PR validation:

please refer to #36009

if this PR is a backport please specify the original PR and why you need to backport that PR:

This is a backport of #36009 to 12_1_X.
The reason of this PR is to backport the fixes concerning critical issues found in L2NNTag modules and deepTauID for HLT.

[cmsbuild]

A new Pull Request was created by @azotz for CMSSW_12_1_X.

It involves the following packages:

• RecoTauTag/HLTProducers (hlt)
• RecoTauTag/RecoTau (reconstruction)

@jpata, @missirol, @cmsbuild, @Martin-Grunewald, @slava77 can you please review it and eventually sign? Thanks.
@silviodonato, @mbluj, @azotz this is something you requested to watch as well.
@perrotta, @dpiparo, @qliphy you are the release manager for this.

cms-bot commands are listed here

• Backported from Fixes to L2Tau CNN and deepTauID for HLT #36009

[missirol]

In my understanding, the L2TauTagFilter producer is not used in any central workflow, so the physics change in it for high-pT taus should not affect any outputs. I guess this makes it okay to backport these updates.

The other changes are limited to fillDescriptions methods.

[missirol]

please test

[cmsbuild]

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-efbfc7/20460/summary.html
COMMIT: 22c4707
CMSSW: CMSSW_12_1_X_2021-11-10-2300/slc7_amd64_gcc900
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/36089/20460/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

• No significant changes to the logs found
• Reco comparison results: 16 differences found in the comparisons
• DQMHistoTests: Total files compared: 42
• DQMHistoTests: Total histograms compared: 2901440
• DQMHistoTests: Total failures: 0
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 2901418
• DQMHistoTests: Total skipped: 22
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 41 files compared)
• Checked 177 log files, 37 edm output root files, 42 DQM output files
• TriggerResults: no differences found

[slava77]

backport of #36009

[slava77]

+reconstruction

for #36089 22c4707

• code changes are the same as in Fixes to L2Tau CNN and deepTauID for HLT #36009 , which was merged in CMSSW_12_2_X_2021-11-10-1100
• jenkins tests pass and comparisons with the baseline show no (relevant) differences
  • the differences appearing in the reco electron plots are only in mvaOutput().dnn_e_bkgPhoton variable, which is not initialized; the fix to that is in fix memory bug in DNN-based ID (packport of #35984) #36091

[missirol]

+hlt

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next CMSSW_12_1_X IBs (tests are also fine) and once validation in the development release cycle CMSSW_12_2_X is complete. This pull request will now be reviewed by the release team before it's merged. @perrotta, @dpiparo, @qliphy (and backports should be raised in the release meeting by the corresponding L2)

[qliphy]

+1