-
Notifications
You must be signed in to change notification settings - Fork 34
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
Technical update to the DeepJet and DeepDoubleX TF models #27
Conversation
A new Pull Request was created by @hqucms (Huilin Qu) for branch master. @perrotta, @smuzaffar, @mrodozov, @cmsbuild, @slava77, @tulamor can you please review it and eventually sign? Thanks. |
@cmsbuild please test |
The tests are being triggered in jenkins. |
The tests are being triggered in jenkins.
|
+1 |
Comparison job queued. |
-1 Tested at: 9f5e916 CMSSW: CMSSW_11_1_X_2020-03-09-2300 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: You can see the results of the tests here: I found follow errors while testing this PR Failed tests: RelVals AddOn
When I ran the RelVals I found an error in the following workflows: runTheMatrix-results/140.53_RunHI2011+RunHI2011+RECOHID11+HARVESTDHI/step2_RunHI2011+RunHI2011+RECOHID11+HARVESTDHI.log5.1 step1 runTheMatrix-results/5.1_TTbar+TTbarFS+HARVESTFS/step1_TTbar+TTbarFS+HARVESTFS.log135.4 step1 runTheMatrix-results/135.4_ZEE_13+ZEEFS_13+HARVESTUP15FS+MINIAODMCUP15FS/step1_ZEE_13+ZEEFS_13+HARVESTUP15FS+MINIAODMCUP15FS.log1001.0 step2 runTheMatrix-results/1001.0_RunMinBias2011A+RunMinBias2011A+TIER0EXP+ALCAEXP+ALCAHARVDSIPIXELCALRUN1+ALCAHARVD1+ALCAHARVD2+ALCAHARVD3+ALCAHARVD4+ALCAHARVD5/step2_RunMinBias2011A+RunMinBias2011A+TIER0EXP+ALCAEXP+ALCAHARVDSIPIXELCALRUN1+ALCAHARVD1+ALCAHARVD2+ALCAHARVD3+ALCAHARVD4+ALCAHARVD5.log1000.0 step2 runTheMatrix-results/1000.0_RunMinBias2011A+RunMinBias2011A+TIER0+SKIMD+HARVESTDfst2+ALCASPLIT/step2_RunMinBias2011A+RunMinBias2011A+TIER0+SKIMD+HARVESTDfst2+ALCASPLIT.log140.56 step2 runTheMatrix-results/140.56_RunHI2018+RunHI2018+RECOHID18+HARVESTDHI18/step2_RunHI2018+RunHI2018+RECOHID18+HARVESTDHI18.log4.53 step3 runTheMatrix-results/4.53_RunPhoton2012B+RunPhoton2012B+HLTD+RECODR1reHLT+HARVESTDR1reHLT/step3_RunPhoton2012B+RunPhoton2012B+HLTD+RECODR1reHLT+HARVESTDR1reHLT.log136.731 step3 runTheMatrix-results/136.731_RunSinglePh2016B+RunSinglePh2016B+HLTDR2_2016+RECODR2_2016reHLT_skimSinglePh_HIPM+HARVESTDR2/step3_RunSinglePh2016B+RunSinglePh2016B+HLTDR2_2016+RECODR2_2016reHLT_skimSinglePh_HIPM+HARVESTDR2.log158.0 step2 runTheMatrix-results/158.0_HydjetQ_B12_5020GeV_2018_ppReco+HydjetQ_B12_5020GeV_2018_ppReco+DIGIHI2018PPRECO+RECOHI2018PPRECO+ALCARECOHI2018PPRECO+HARVESTHI2018PPRECO/step2_HydjetQ_B12_5020GeV_2018_ppReco+HydjetQ_B12_5020GeV_2018_ppReco+DIGIHI2018PPRECO+RECOHI2018PPRECO+ALCARECOHI2018PPRECO+HARVESTHI2018PPRECO.log136.793 step3 runTheMatrix-results/136.793_RunDoubleEG2017C+RunDoubleEG2017C+HLTDR2_2017+RECODR2_2017reHLT_skimDoubleEG_Prompt+HARVEST2017/step3_RunDoubleEG2017C+RunDoubleEG2017C+HLTDR2_2017+RECODR2_2017reHLT_skimDoubleEG_Prompt+HARVEST2017.log136.874 step3 runTheMatrix-results/136.874_RunEGamma2018C+RunEGamma2018C+HLTDR2_2018+RECODR2_2018reHLT_skimEGamma_Offline_L1TEgDQM+HARVEST2018_L1TEgDQM/step3_RunEGamma2018C+RunEGamma2018C+HLTDR2_2018+RECODR2_2018reHLT_skimEGamma_Offline_L1TEgDQM+HARVEST2018_L1TEgDQM.log1330.0 step3 runTheMatrix-results/1330.0_ZMM_13+ZMM_13+DIGIUP15+RECOUP15_L1TMuDQM+HARVESTUP15_L1TMuDQM+NANOUP15/step3_ZMM_13+ZMM_13+DIGIUP15+RECOUP15_L1TMuDQM+HARVESTUP15_L1TMuDQM+NANOUP15.log1306.0 step3 runTheMatrix-results/1306.0_SingleMuPt1_UP15+SingleMuPt1_UP15+DIGIUP15+RECOUP15+HARVESTUP15/step3_SingleMuPt1_UP15+SingleMuPt1_UP15+DIGIUP15+RECOUP15+HARVESTUP15.log10042.0 step3 runTheMatrix-results/10042.0_ZMM_13+ZMM_13TeV_TuneCUETP8M1_2017_GenSimFull+DigiFull_2017+RecoFull_2017+HARVESTFull_2017+ALCAFull_2017+NanoFull_2017/step3_ZMM_13+ZMM_13TeV_TuneCUETP8M1_2017_GenSimFull+DigiFull_2017+RecoFull_2017+HARVESTFull_2017+ALCAFull_2017+NanoFull_2017.log9.0 step3 runTheMatrix-results/9.0_Higgs200ChargedTaus+Higgs200ChargedTaus+DIGI+RECO+HARVEST/step3_Higgs200ChargedTaus+Higgs200ChargedTaus+DIGI+RECO+HARVEST.log11634.0 step2 runTheMatrix-results/11634.0_TTbar_14TeV+TTbar_14TeV_TuneCP5_2021_GenSimFull+DigiFull_2021+RecoFull_2021+HARVESTFull_2021+ALCAFull_2021/step2_TTbar_14TeV+TTbar_14TeV_TuneCP5_2021_GenSimFull+DigiFull_2021+RecoFull_2021+HARVESTFull_2021+ALCAFull_2021.log12434.0 step2 runTheMatrix-results/12434.0_TTbar_14TeV+TTbar_14TeV_TuneCP5_2023_GenSimFull+DigiFull_2023+RecoFull_2023+HARVESTFull_2023+ALCAFull_2023/step2_TTbar_14TeV+TTbar_14TeV_TuneCP5_2023_GenSimFull+DigiFull_2023+RecoFull_2023+HARVESTFull_2023+ALCAFull_2023.log25.0 step3 runTheMatrix-results/25.0_TTbar+TTbar+DIGI+RECOAlCaCalo+HARVEST+ALCATT/step3_TTbar+TTbar+DIGI+RECOAlCaCalo+HARVEST+ALCATT.log10824.0 step3 runTheMatrix-results/10824.0_TTbar_13+TTbar_13TeV_TuneCUETP8M1_2018_GenSimFull+DigiFull_2018+RecoFull_2018+HARVESTFull_2018+ALCAFull_2018+NanoFull_2018/step3_TTbar_13+TTbar_13TeV_TuneCUETP8M1_2018_GenSimFull+DigiFull_2018+RecoFull_2018+HARVESTFull_2018+ALCAFull_2018+NanoFull_2018.log10024.0 step3 runTheMatrix-results/10024.0_TTbar_13+TTbar_13TeV_TuneCUETP8M1_2017_GenSimFull+DigiFull_2017+RecoFull_2017+HARVESTFull_2017+ALCAFull_2017+NanoFull_2017/step3_TTbar_13+TTbar_13TeV_TuneCUETP8M1_2017_GenSimFull+DigiFull_2017+RecoFull_2017+HARVESTFull_2017+ALCAFull_2017+NanoFull_2017.log25202.0 step3 runTheMatrix-results/25202.0_TTbar_13+TTbar_13+DIGIUP15_PU25+RECOUP15_PU25+HARVESTUP15_PU25+NANOUP15_PU25/step3_TTbar_13+TTbar_13+DIGIUP15_PU25+RECOUP15_PU25+HARVESTUP15_PU25+NANOUP15_PU25.log10224.0 step3 runTheMatrix-results/10224.0_TTbar_13+TTbar_13TeV_TuneCUETP8M1_2017PU_GenSimFull+DigiFullPU_2017PU+RecoFullPU_2017PU+HARVESTFullPU_2017PU+NanoFull_2017PU/step3_TTbar_13+TTbar_13TeV_TuneCUETP8M1_2017PU_GenSimFull+DigiFullPU_2017PU+RecoFullPU_2017PU+HARVESTFullPU_2017PU+NanoFull_2017PU.log20034.0 step3 runTheMatrix-results/20034.0_TTbar_14TeV+TTbar_14TeV_TuneCP5_2026D35_GenSimHLBeamSpotFull14+DigiFullTrigger_2026D35+RecoFullGlobal_2026D35+HARVESTFullGlobal_2026D35/step3_TTbar_14TeV+TTbar_14TeV_TuneCP5_2026D35_GenSimHLBeamSpotFull14+DigiFullTrigger_2026D35+RecoFullGlobal_2026D35+HARVESTFullGlobal_2026D35.log20434.0 step3 runTheMatrix-results/20434.0_TTbar_14TeV+TTbar_14TeV_TuneCP5_2026D41_GenSimHLBeamSpotFull14+DigiFullTrigger_2026D41+RecoFullGlobal_2026D41+HARVESTFullGlobal_2026D41/step3_TTbar_14TeV+TTbar_14TeV_TuneCP5_2026D41_GenSimHLBeamSpotFull14+DigiFullTrigger_2026D41+RecoFullGlobal_2026D41+HARVESTFullGlobal_2026D41.log21234.0 step3 runTheMatrix-results/21234.0_TTbar_14TeV+TTbar_14TeV_TuneCP5_2026D44_GenSimHLBeamSpotFull14+DigiFullTrigger_2026D44+RecoFullGlobal_2026D44+HARVESTFullGlobal_2026D44/step3_TTbar_14TeV+TTbar_14TeV_TuneCP5_2026D44_GenSimHLBeamSpotFull14+DigiFullTrigger_2026D44+RecoFullGlobal_2026D44+HARVESTFullGlobal_2026D44.log23234.0 step3 runTheMatrix-results/23234.0_TTbar_14TeV+TTbar_14TeV_TuneCP5_2026D49_GenSimHLBeamSpotFull14+DigiFullTrigger_2026D49+RecoFullGlobal_2026D49+HARVESTFullGlobal_2026D49/step3_TTbar_14TeV+TTbar_14TeV_TuneCP5_2026D49_GenSimHLBeamSpotFull14+DigiFullTrigger_2026D49+RecoFullGlobal_2026D49+HARVESTFullGlobal_2026D49.log250202.181 step4 runTheMatrix-results/250202.181_TTbar_13UP18+TTbar_13UP18+PREMIXUP18_PU25+DIGIPRMXLOCALUP18_PU25+RECOPRMXUP18_PU25+HARVESTUP18_PU25/step4_TTbar_13UP18+TTbar_13UP18+PREMIXUP18_PU25+DIGIPRMXLOCALUP18_PU25+RECOPRMXUP18_PU25+HARVESTUP18_PU25.log
I found errors in the following addon tests: cmsDriver.py TTbar_8TeV_TuneCUETP8M1_cfi --conditions auto:run1_mc --fast -n 100 --eventcontent AODSIM,DQM --relval 100000,1000 -s GEN,SIM,RECOBEFMIX,DIGI:pdigi_valid,L1,DIGI2RAW,L1Reco,RECO,EI,VALIDATION --customise=HLTrigger/Configuration/CustomConfigs.L1THLT --datatier GEN-SIM-DIGI-RECO,DQMIO --beamspot Realistic8TeVCollision : FAILED - time: date Tue Mar 10 23:26:50 2020-date Tue Mar 10 23:24:22 2020 s - exit: 16896 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: |
Comparison not run due to runTheMatrix errors (RelVals and Igprof tests were also skipped) |
The tests are being triggered in jenkins. |
Comparison is ready Comparison Summary:
|
Comparison is ready Comparison Summary:
|
The tests are being triggered in jenkins. |
+1 |
Comparison job queued. |
Comparison is ready Comparison Summary:
|
The tests are being triggered in jenkins. |
+1 |
Comparison job queued. |
Comparison is ready Comparison Summary:
|
@hqucms , should we close this PR ( cms-sw/cmssw#29172 was closed too). |
@smuzaffar Yes this is no longer needed -- let me close it. |
A few technical updates to the TF models (e.g., remove nodes that are only used for training like
cpf_input_batchnorm/keras_learning_phase
from the graph, fix the DeepDoubleX models for batch inference). No change in the output is expected. Small (5-10%) speedup and slightly reduced model size.The models are converted from the original Keras models using
https://github.com/hqucms/keras_to_tensorflow/blob/master/keras_to_tensorflow.py
[Updated on backward compatibility]
These models are intended to be used together w/ cms-sw/cmssw#29172, though it is OK to merge this PR (the DNN models) before cms-sw/cmssw#29172, because currently the inference in CMSSW is using ONNX (since cms-sw/cmssw#28112), so these TF models are not really used.
These models are NOT compatible w/ older releases (before cms-sw/cmssw#28112) -- one needs to remove the
learning_phase
input in https://github.com/cms-sw/cmssw/blob/8a7ee7592d72cc40cc0a95c371787ccc49f520b1/RecoBTag/TensorFlow/plugins/DeepDoubleXTFJetTagsProducer.cc#L136 and https://github.com/cms-sw/cmssw/blob/8a7ee7592d72cc40cc0a95c371787ccc49f520b1/RecoBTag/TensorFlow/plugins/DeepFlavourTFJetTagsProducer.cc#L141 to use these new models.