search

cms-analysis / flashgg

20 stars 158 forks source link

Fix the bug which cannot extract the PDF for RunII UL signal samples (Summer20) #1285

Closed youyingli closed 2 years ago

youyingli commented 2 years ago

The PR follows #1268 and has a more general syntax. Most of the madgraph samples in Summer20UL18 (at least H->gg samples) are assumed with a new PDF recipe called "NNPDF31_nnlo_as_0118_mc_hessian_pdfas" with id = 325300. The information is given by LHAPDF as the following :

<weightgroup combine="symmhessian+as" name="NNPDF31_nnlo_as_0118_mc_hessian_pdfas"> # 325300: NNPDF3.1 NNLO global fit, alphas(MZ)=0.118. mem=0 ; average on replicas (setting negative replicas to zero); mem=1-100 ; PDF eig.; mem=101 ; central value (forced positive definite) Alphas(MZ)=0.116; mem=102 ; central value (forced positive definite) Alphas(MZ)=0.120

Since it is symmhessian, maybe the additional hessian calculation is no long nesserary?

As for Powheg is assigned as "NNPDF31_nnlo_hessian_pdfas" with id = 306000.

The PR was tested using Powheg and Madgraph generator (the majority of the Higgs boson production) for Summer20UL18 samples. All can work well.

#Powheg
cmsRun MicroAOD/test/microAODstd.py processType=sig datasetName=vh conditionsJSON=MetaData/data/MetaConditions/Era2018_legacy_v1.json fileNames=/store/mc/RunIISummer20UL18MiniAODv2/ZH_HToGG_ZToAll_M125_TuneCP5_13TeV-powheg-pythia8/MINIAODSIM/106X_upgrade2018_realistic_v16_L1v1-v1/2430000/088AAA94-0467-8842-9EEE-FA55830E7A80.root

#Madgraph
cmsRun MicroAOD/test/microAODstd.py processType=sig datasetName=ggh conditionsJSON=MetaData/data/MetaConditions/Era2018_legacy_v1.json fileNames=/store/mc/RunIISummer20UL18MiniAODv2/GluGluHToGG_M125_TuneCP5_13TeV-amcatnloFXFX-pythia8/MINIAODSIM/106X_upgrade2018_realistic_v16_L1v1-v1/2430000/077C0DBB-28B8-7E4E-A7EB-5640F154B83B.root