New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Update miniAOD AK8 userFloat content. Add PUPPI substructure. Remove CMSTT. #13793
Conversation
A new Pull Request was created by @jdolen (James Dolen) for CMSSW_8_1_X. It involves the following packages: CommonTools/RecoAlgos @cvuosalo, @monttj, @cmsbuild, @slava77, @vadler, @davidlange6 can you please review it and eventually sign? Thanks. cms-bot commands are list here #13028 |
#'keep *_slimmedJetsAK8PFCHSSoftDropSubjets_*_*', | ||
#'keep *_slimmedJetsCMSTopTagCHSSubjets_*_*', | ||
'keep *_slimmedJetsAK8PFCHSSoftDropPacked_SubJets_*', | ||
'keep *_slimmedJetsCMSTopTagCHSPacked_SubJets_*', | ||
'keep *_slimmedJetsAK8PFPuppiSoftDropPacked_SubJets_*', | ||
#'keep *_slimmedJetsCMSTopTagCHSPacked_SubJets_*', |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
please drop commented out lines,
unless there is a good reason to keep.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Also remove 'keep *_caTopTagInfosPAT_*_*'
in line 65.
@cmsbuild please test |
The tests are being triggered in jenkins. |
|
||
# Add AK8 top tagging variables | ||
process.patJetsAK8.tagInfoSources = cms.VInputTag(cms.InputTag("caTopTagInfosPAT")) | ||
process.patJetsAK8.userData.userFloats.src += ['ak8PFJetsCHSPrunedMass','ak8PFJetsCHSSoftDropMass'] #,'ak8PFJetsCHSTrimmedMass','ak8PFJetsCHSFilteredMass'] | ||
process.patJetsAK8.addTagInfos = cms.bool(True) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
process.patJetsAK8.addTagInfos
can now be set to False
or, since that is the default value for this switch, this line can be removed.
An extended test of workflow 25202.0_TTbar_13 with 70 events and no DQM/Validation step against baseline CMSSW_8_1_X_2016-03-14-1100 shows the following time and memory usage. TopTagCHS modules are removed and replaced by AK8PFPuppi modules. The removed modules had a total time of about 50 ms/evt, while the new modules together only take about 40 ms/evt. Overall, there is no significant change in CPU time for RECO.
|
Pull request #13793 was updated. @cmsbuild, @cvuosalo, @slava77, @davidlange6 can you please check and sign again. |
Pull request #13793 was updated. @cmsbuild, @cvuosalo, @slava77, @davidlange6 can you please check and sign again. |
Pull request #13793 was updated. @cmsbuild, @cvuosalo, @slava77, @davidlange6 can you please check and sign again. |
Pull request #13793 was updated. @cmsbuild, @cvuosalo, @slava77, @davidlange6 can you please check and sign again. |
@cmsbuild please test |
The tests are being triggered in jenkins. |
-1 Tested at: 8201e22 ---> test runtestPhysicsToolsPatAlgos had ERRORS you can see the results of the tests here: |
Additional, possibly more accurate, test results running only the PAT step with 1000 events to produce Mini-AOD were obtained with this command:
Wall-clock time and memory usage are shown, along with a second timing measure that excludes the first event. Only the largest time changes are listed (see above for full listing of added and removed modules).
Change in the size of Mini-AOD:
|
+1 Removing obsolete and unused CMS Top taggers and adding PUPPI-based taggers to Mini-AOD. #13796 is the 80X version of this PR, and it has already been approved. The code changes are satisfactory. Jenkins tests for this PR encountered errors in the IB, but Jenkins tests for #13796 show no significant differences, except that DQM plots for the removed taggers are now empty. Extended tests are discussed above in detail, and they show no problems. |
We have the following proposal for the AK8chs jet MiniAOD userfloat content in 80X/81X:
Size/timing estimates are given in the slides linked here.
https://indico.cern.ch/event/510570/contribution/7/attachments/1246782/1836484/160321_-_Jet_substructure_MiniAOD_content.pptx.pdf
Cheers,
@rappoccio @ahinzmann @jdolen