Example on checking xAODs in python
This document was originally written by Dr. Shawfeng Dong
This repo contains a few examples for processing ATLAS xAOD data.
What information is in an xAOD file
The ATLAS Analysis Release provides a handy Python checkxAOD.py that we can use to learn the container types and container keys in an xAOD file.
First, set up the environment for ATLAS Analysis
export ATLAS_LOCAL_ROOT_BASE=/cvmfs/atlas.cern.ch/repo/ATLASLocalRootBase
source $ATLAS_LOCAL_ROOT_BASE/user/atlasLocalSetup.sh
asetup AnalysisBase,21.2.111
Then run checkxAOD.py on a sample xAOD file:
[shawfeng@cent7a ATLAS]$ which checkxAOD.py
/cvmfs/atlas.cern.ch/repo/sw/software/21.2/AnalysisBase/21.2.111/InstallArea/x86_64-centos7-gcc8-opt/bin/checkxAOD.py
[shawfeng@cent7a ATLAS]$ checkxAOD.py DAOD_SUSY15.11525262._000003.pool.root.1
========================================================================================================================
File: DAOD_SUSY15.11525262._000003.pool.root.1
------------------------------------------------------------------------------------------------------------------------
Memory size Disk Size Size/Event Compression Entries Name (Type)
------------------------------------------------------------------------------------------------------------------------
14.21 kB 3.50 kB 0.00 kB/event 4.06 855 TrigConfKeys (xAOD::TrigConfKeys_v1) [Trig]
17.60 kB 4.03 kB 0.00 kB/event 4.36 855 AntiKt4LCTopoJets (DataVector<xAOD::Jet_v1>) [Jet]
71.30 kB 11.11 kB 0.01 kB/event 6.42 855 Kt4EMTopoEventShape (xAOD::EventShape_v1) [PFO]
103.82 kB 17.25 kB 0.02 kB/event 6.02 855 HLT_xAOD__TrackParticleContainer_InDetTrigTrackingxAODCnv_Muon_EFID (DataVector<xAOD::TrackParticle_v1>) [Trig]
158.71 kB 26.15 kB 0.03 kB/event 6.07 855 LVL1EnergySumRoI (xAOD::EnergySumRoI_v1) [Trig]
141.89 kB 38.54 kB 0.05 kB/event 3.68 855 LVL1MuonRoIs (DataVector<xAOD::MuonRoI_v1>) [Trig]
165.77 kB 47.94 kB 0.06 kB/event 3.46 855 MET_LocHadTopo (xAOD::MissingETContainer_v1) [MET]
151.53 kB 49.34 kB 0.06 kB/event 3.07 855 HLT_xAOD__TrigMissingETContainer_TrigEFMissingET_topocl_PUC (DataVector<xAOD::TrigMissingET_v1>) [Trig]
150.48 kB 49.71 kB 0.06 kB/event 3.03 855 HLT_xAOD__TrigMissingETContainer_TrigEFMissingET (DataVector<xAOD::TrigMissingET_v1>) [Trig]
150.86 kB 50.13 kB 0.06 kB/event 3.01 855 HLT_xAOD__TrigMissingETContainer_TrigEFMissingET_mht (DataVector<xAOD::TrigMissingET_v1>) [Trig]
132.85 kB 51.15 kB 0.06 kB/event 2.60 855 HLT_xAOD__ElectronContainer_egamma_Electrons (DataVector<xAOD::Electron_v1>) [Trig]
319.77 kB 63.15 kB 0.07 kB/event 5.06 855 EventInfo (xAOD::EventInfo_v1) [EvtId]
242.30 kB 68.48 kB 0.08 kB/event 3.54 855 HLT_xAOD__MuonContainer_MuonEFInfo_FullScan (DataVector<xAOD::Muon_v1>) [Trig]
179.61 kB 78.31 kB 0.09 kB/event 2.29 855 HLT_xAOD__PhotonContainer_egamma_Photons (DataVector<xAOD::Photon_v1>) [Trig]
242.57 kB 82.72 kB 0.10 kB/event 2.93 855 MET_Core_AntiKt4EMTopo (xAOD::MissingETContainer_v1) [MET]
359.62 kB 85.67 kB 0.10 kB/event 4.20 855 HLT_xAOD__MuonContainer_MuonEFInfo (DataVector<xAOD::Muon_v1>) [Trig]
670.42 kB 109.75 kB 0.13 kB/event 6.11 855 ByteStreamEventInfo (EventInfo_p4) [EvtId]
365.98 kB 133.39 kB 0.16 kB/event 2.74 855 GSFConversionVertices (DataVector<xAOD::Vertex_v1>) [egamma]
279.50 kB 136.51 kB 0.16 kB/event 2.05 855 MuonSpectrometerTrackParticles (DataVector<xAOD::TrackParticle_v1>) [Muon]
430.51 kB 138.66 kB 0.16 kB/event 3.10 855 MET_Reference_AntiKt4EMTopo (xAOD::MissingETContainer_v1) [MET]
466.98 kB 142.38 kB 0.17 kB/event 3.28 855 VrtSecInclusive_SecondaryVertices (DataVector<xAOD::Vertex_v1>) [*Unknown*]
1002.16 kB 202.54 kB 0.24 kB/event 4.95 855 LVL1EmTauRoIs (DataVector<xAOD::EmTauRoI_v2>) [Trig]
476.07 kB 208.92 kB 0.24 kB/event 2.28 855 ExtrapolatedMuonTrackParticles (DataVector<xAOD::TrackParticle_v1>) [Muon]
8911.50 kB 236.74 kB 0.28 kB/event 37.64 855 xTrigDecision (xAOD::TrigDecision_v1) [Trig]
511.68 kB 240.21 kB 0.28 kB/event 2.13 855 CombinedMuonTrackParticles (DataVector<xAOD::TrackParticle_v1>) [Muon]
638.75 kB 311.45 kB 0.36 kB/event 2.05 855 MET_Track (xAOD::MissingETContainer_v1) [MET]
5424.75 kB 566.53 kB 0.66 kB/event 9.58 855 TauJets (DataVector<xAOD::TauJet_v2>) [tau]
1486.02 kB 583.31 kB 0.68 kB/event 2.55 855 Muons (DataVector<xAOD::Muon_v1>) [Muon]
1001.48 kB 622.72 kB 0.73 kB/event 1.61 855 egammaClusters (DataVector<xAOD::CaloCluster_v1>) [egamma]
2177.31 kB 851.52 kB 1.00 kB/event 2.56 855 METAssoc_AntiKt4EMTopo (xAOD::MissingETAssociationMap_v1) [MET]
1240.53 kB 861.32 kB 1.01 kB/event 1.44 855 egammaTopoSeededClusters (DataVector<xAOD::CaloCluster_v1>) [egamma]
2268.96 kB 919.86 kB 1.08 kB/event 2.47 855 MuonSegments (DataVector<xAOD::MuonSegment_v1>) [Muon]
1288.68 kB 936.54 kB 1.10 kB/event 1.38 855 InDetForwardTrackParticles (DataVector<xAOD::TrackParticle_v1>) [InDet]
11424.90 kB 1035.51 kB 1.21 kB/event 11.03 855 BTagging_AntiKt4EMTopo (DataVector<xAOD::BTagging_v1>) [BTag]
2550.01 kB 1252.15 kB 1.46 kB/event 2.04 855 Electrons (DataVector<xAOD::Electron_v1>) [egamma]
8677.78 kB 1786.35 kB 2.09 kB/event 4.86 855 PrimaryVertices (DataVector<xAOD::Vertex_v1>) [InDet]
9377.16 kB 1989.75 kB 2.33 kB/event 4.71 855 AntiKt4EMTopoJets (DataVector<xAOD::Jet_v1>) [Jet]
4405.27 kB 2562.17 kB 3.00 kB/event 1.72 855 GSFTrackParticles (DataVector<xAOD::TrackParticle_v1>) [egamma]
4534.35 kB 2806.48 kB 3.28 kB/event 1.62 855 Photons (DataVector<xAOD::Photon_v1>) [egamma]
12708.16 kB 7513.35 kB 8.79 kB/event 1.69 855 VrtSecInclusive_SelectedTrackParticles (DataVector<xAOD::TrackParticle_v1>) [*Unknown*]
37973.60 kB 9237.21 kB 10.80 kB/event 4.11 855 VrtSecInclusive_All2TrksVertices (DataVector<xAOD::Vertex_v1>) [*Unknown*]
40923.05 kB 11891.41 kB 13.91 kB/event 3.44 855 TrigNavigation (xAOD::TrigNavigation_v1) [Trig]
35664.74 kB 21271.48 kB 24.88 kB/event 1.68 855 InDetTrackParticles (DataVector<xAOD::TrackParticle_v1>) [InDet]
------------------------------------------------------------------------------------------------------------------------
199483.14 kB 69275.38 kB 81.02 kB/event Total
========================================================================================================================
================================================================================
Categorized data
================================================================================
Disk Size Fraction Category Name
--------------------------------------------------------------------------------
0.013 kb 0.000 PFO
0.202 kb 0.002 EvtId
0.663 kb 0.008 tau
1.211 kb 0.015 BTag
1.675 kb 0.021 MET
2.332 kb 0.029 Jet
2.443 kb 0.030 Muon
9.635 kb 0.119 egamma
15.028 kb 0.185 Trig
19.758 kb 0.244 *Unknown*
28.064 kb 0.346 InDet
81.024 kb 1.000 Total
================================================================================
CSV for categories disk size/evt and fraction:
Total,InDet,*Unknown*,Trig,egamma,Muon,Jet,MET,BTag,tau,EvtId,PFO
81.024,28.064,19.758,15.028,9.635,2.443,2.332,1.675,1.211,0.663,0.202,0.013
1.000,0.346,0.244,0.185,0.119,0.030,0.029,0.021,0.015,0.008,0.002,0.000
================================================================================
In this case, for example, AntiKt4LCTopoJets is a key name for the container whose type is DataVector<xAOD::Jet_v1>. Once you know the container types, you can find their APIs at the RootCore APIs site.
Using xAOD API in python
In your analysis, you choose the key name for the particular instance of the container in which you are interested (e.g., AntiKt4LCTopoJets). Note xAOD APIs are higher-level APIs, and they hide some behind-the-scenes magic, which is actually saved in the Auxiliary store of the raw Root file. However, as of this writing, you can only access the xAOD APIs in either C++ (e.g., Exam_JetsPlot.cxx) or pyROOT (e.g., pyROOT_example.ipynb, but not in uproot yet. Until someone implements xAOD support for uproot, you will have to read the raw data from the Auxiliary store (for an example, see convert_specific_variables.py).