General Instructions:

https://twiki.cern.ch/twiki/bin/view/CMS/MuonOfflineDQM Many plots are not clearly visible: titles, axis labels, cannot read entries, mean, RMS , titles unclear. Not reference plots for every distribution. Unclear what is real difference from reference and what to do in that case.

Run 110756, has 34 LS, ExpressMuon/CRAFT09-Express-v1/FEVT

In general: display overflow, underflow? Errors on mean, RMS?

how to treat quantities dependent on calibrations, aligments? will get better with reprocessing even in not correct in prompt offline? even if maybe this is after first Alca reco constants computed...

MuonEnergyDepositAnalyzer

In Instructions: Energy deposits (separately for ECAL/HCAL/HO subdetectors and Barrel/EndCap regions) Main histograms to check: ECAL+HCAL S9 energy deposits in the barrel region

Code: CMSSW/DQMOffline/Muon/src/MuonEnergyDepositAnalyzer.h. DQM monitoring source for muon energy deposits. Takes as input "muons" collection.

Set many histos to log scale by default! Maybe cut on the range. I think one could keep only few histos from here! Do we need both barrel and endcap? Do we need specific histos for pointing muons??

  • ecalDepositedEnergyBarrel_muons. Energy deposited in the ECAL barrel cells. Entries 238966 (is this number of "muons"? Or "Nmuons" times "Ncells"?) Is this one that should have a peak at 200 MeV?? Then to see it properly consider changing range. Should be log!
  • ecalDepositedEnergyEndcap_muons. Energy deposited in the ECAL endcap cells. Entries 4013 (is this number of muons? Number of muons in endcap region? NMuonsEncap*NCellsEndcap???)
  • ecalS9DepositedEnergyBarrel_muons. Energy deposited in the ECAL barrel 3*3 towers. Entries 238966 (?) No maximum observed at ~200 MeV as said in instructions...
  • ecalS9DepositedEnergyBarrel_muons. Energy deposited in the ECAL endcap 3*3 towers. Entries 4013

  • ecalS9DepositedEnergy_Glb,Sta,Tk_muons. For pointing glb (880),sta(1073), trk(131) muons energy deposited in ECAL 3*3 towers. Sum of energy in barrel and endcap??

  • Same for HCAL. The peaks that should be visible in 3*3 histos (suggested in the instructions) can only be seen in the general plots (not 3*3 towers, but all barrel/endcap) Entries 237685,5294,237685,5294,880,1073,131

  • Same for HO (w/o division barrel endcap, of course) Entries 235066, 235066, 880,1073,131

Are peaks at zero due to acceptance where muons do not cross the calorimeters? Peaks much reduced for pointing muons...

MuonRecoAnalyzer

In Instructions: Main histograms to check: Kinematics distributions (radial and axial direction, transverse momentum, pseudorapity, charge) for Glb-STA-TK muons, Residuals between the 3 glb muon tracks, RecHits provenance, Muon tracks provenance.

Code: /CMSSW/DQMOffline/Muon/src/MuonRecoAnalyzer.h. DQM monitoring source for muon reco track

To my opinion there are too many plots here!!!

  • GlbMuon_Glb_chi2OverDf: Chi2 OverDF_GLB. Entries 12559 (must be numb glb muons). Range 0,50. Consider log scale? Mean 1291!!! RNS 4.8E+04
  • GlbMuon_Glb_eta: Eta Glb. Entries 12559 (OK, nglb) Range -3,3. Mean ~0.1, RMS ~0.5 Asymmetric, 2 peaks, bigger at ~0.5.
  • GlbMuon_Glb_p: p Glb. Entries 12559 (OK, nglb) Range 0,500. Peak ~10 GeV. Mean ~ 50 GeV, RMS ~150
  • GlbMuon_Glb_phi: phi Glb. Entries 12559 (OK, nglb). Range -3,3. Very asymm, concentrated at negative phi. Mean ~-1.6, RMS ~ 0.8
  • GlbMuon_Glb_pt: pt GLB. Entries 12559 (OK, nglb). Range 0,500. Mean ~ 40 GeV, RMS ~120
  • GlbMuon_Glb_q. q GLB. Entries 12559 (OK, nglb). Range -2.5,2.5. More positive muons. Mean ~0.11, RMS ~1.
  • GlbMuon_Glb_theta. Theta GLB. Entries 12559 (OK, nglb). Range -3,3. Mean ~1.5, RMS ~0.5

Same for STA and TK from GLB. All have entries equal to previous histos (numb of glb muons)

Extrange things comparing with GLB fit... For GLB fit:

For STA component:
  • GlbMuon_STA_chi2OverDf has Mean ~3, RMS ~7.
  • GlbMuon_Glb_p has Mean ~100, RMS ~4000.
  • GlbMuon_Glb_pt Has Mean ~90, RMS~4000.
For TK component:
  • GlbMuon_TK_chi2OverDf: Mean ~1, RMS ~0.6.
  • GlbMuon_TK_p Mean ~65, RMS ~900.
  • GlbMuon_TK_pt Mean ~56, RMS ~850.

* GlbMuon _qComparison. comparison between qGLB abd qTKfromGLB,qSTAfromGLB. Labels not correctly displayed. * GlbRh _Div_RhAssStaTK: recHitsGLB/(recHits_TKfromGLB+recHits_STAfromGLB) Range 0,1. Big peak at 1. Consider log scale?

* ResVsEta _GlbSta_eta. (Eta_GLB-Eta_STAfromGLB) vs Eta_GLB * ResVsEta _GlbSta_oneOverpt. (Eta_GLB-Eta_STAfromGLB) vs (1/pt)_GLB WRONG!! Should be ((1/pt)_GLB-(1/pt)_STAfromGLB) vs Eta_GLB. It is oneOverptResolution[4]

* ResVsEta _TkGlb_eta. (Eta_TkfromGLB-Eta_GLB) vs Eta_GLB * ResVsEta _TkGlb_oneOverpt. (Eta_TkfromGLB-Eta_GLB) vs (1/pt)_GLB WRONG!! Should be ((1/pt)_TkfromGLB-(1/pt)_GLB) vs Eta_GLB. It is oneOverptResolution[3]

* ResVsEta _TkSta_eta. (Eta_TkfromGLB-Eta_STAfromGLB) vs Eta_TKfromGLB * ResVsEta _TkSta_oneOverpt. (Eta_TkfromGLB-Eta_STAfromGLB) vs (1/pt)_GLB WRONG!! Should be ((1/pt)_TkfromGLB-(1/pt)_STAfromGLB) vs Eta_GLB. It is oneOverptResolution[5]

Same problem for ResVsPhi. Why is order changed in display? Mostly OK for ResVsPt, only problem ResVsPt _TkSta_oneOverpt should be vs (1/pt)_GLB, not vs (1/pt)_TKfromGLB Mostly OK for ResVsTheta, only problem ResVsTheta _TkSta_theta should be vs (Theta)_GLB, not vs (Theta)_TKfromGLB

  • Res_GlbSta_eta. Eta_GLB-ETA_STAfromGLB. Entries 12559. Range -0.03,0.03. Mean 0.0005, RMS ~0.20 Not shown in RED even if discrepancy from ref page...
  • Res_GlbSta_oneOverp. (1/p)_GLB-(1/p)_STAfromGLB. Entries 12559. Range -0.01,0.01. Mean 0.14, RMS 8. Shown as RED!! What is right distribution??? Not corresponding plots on reference page.
  • Res_GlbSta_oneOverpt. (1/pt)_GLB-(1/pt)_STAfromGLB. Entries 12559. Range -0.01,0.01. Mean 0.15, RMS 9. Shown as RED!! What is right distribution??? Not corresponding plots on reference page.
  • Res_GlbSta_phi. Phi_GLB-Phi_STAfromGLB. Entries 12559. Range -0.03,0.03. Mean -0.04, RMS 0.7. Shown as RED!! What is right distribution??? Not corresponding plots on reference page.
  • Res_GlbSta_qOverp. (q/p)_GLB-(q/p)_STAfromGLB. Entries 12559. Range -0.01,0.01. Mean 0.07, RMS ~8. Shown as RED!! What is right distribution??? Not corresponding plots on reference page. * Res_GlbSta_qOverpt. (q/pt)_GLB-(q/pt)_STAfromGLB. Entries 12559. Range -0.01,0.01. Mean 0.08, RMS ~9. Shown as RED!! What is right distribution??? Not corresponding plots on reference page. * Res_GlbSta_theta. (theta)_GLB-(theta)_STAfromGLB. Entries 12559. Range -0.03,0.03. Mean 0.0004, RMS ~0.2. Not corresponding plots on reference page.

Same kind of plots for Res_TkGlb_anything. Only NONRED is Res_TkGlb_theta (how can this be OK, and eta be RED??) Range of plots ~ 1/10 smaller than GlbSta plots. In general means and RMSs ~ same as in Res_GlbSta_anything.

Same kind of plots for Res_TkSta_anything. Only NONRED is Res_TkSta_theta (how can this be OK, and eta be RED??) Range of plots same as GlbSta plots. In general means and RMSs smaller than Almost flat distributions for q/p and phi!!!. WRONG 1/p not displayed, but q/p twice!!!!

  • StaEta. Eta_STAfronGLB (WRONG!!! should be Eta_STA) Entries 241400 (number of STA muons, irregardless whether they are or not glb muon as well. Range -3,3. Mean 0.0734, RMS 0.5359. From code is etaEfficiency[0], filled if recoMuon isStandAloneMuon with recoMuon.StandAloneMuon()->eta().
  • StaEta. Eta_STAfronGLB if isGlb=true. (WRONG!!! should be Eta_STA) Entries 12559 (number of glb muons) Same histogram as GlbMuon_Sta_Eta (No difference whether we plot eta_GLB or eta_STA!!! ???? ) From code is etaEfficiency[1], filled if recoMuon isStandAloneMuon and isGLB with recoMuon.StandAloneMuon()->eta().

  • StaMuon_chi2OverDf. chi2OverDF_STA. Entries 228841 (looking at the code, number of STA that are neither GLB or TK, only STA) Range 0,50. Mean 3.12E+06!!!, RMS 5.44 E+08!!! Why is this not RED????
  • StaMuon_eta. Eta_STA. Entries 228841 (looking at the code, number of STA that are neither GLB or TK, only STA) Range -3,3. Mean 0.07, RMS 0.5
  • StaMuon_p. p_STA. Entries 228841 (looking at the code, number of STA that are neither GLB or TK, only STA) Range 0,500. Mean 109, RMS 1.6E04. Big peak at zero!!! Should be log scale??? Why is this not RED???
  • StaMuon_phi. Phi_STA. Entries 228841 (looking at the code, number of STA that are neither GLB or TK, only STA) Range -3,3. Mean -1.4, RMS 0.9 * StaMuon_pt. pt_STA. Entries 228841 (looking at the code, number of STA that are neither GLB or TK, only STA) Range 0,500. Mean 80, RMS 7262. Big peak at zero!!! Should be log scale??? Why is this not RED??? * StaMuon_q. q_STA. Entries 228841 (looking at the code, number of STA that are neither GLB or TK, only STA) * StaMuon_theta. theta_STA. Entries 228841 (looking at the code, number of STA that are neither GLB or TK, only STA) Range 0.3. Mean 1.5 RMS 0.5

  • StaPhi. Phi_STAfromGLB (WRONG!!! should be Phi_STA) Entries 241400 (number of STA muons, irregardless whether they are or not glb muon as well. Range -3,3. Mean -1.446, RMS 0.9578. From code is phiEfficiency[0], filled if recoMuon isStandAloneMuon with recoMuon.StandAloneMuon()->phi().
  • StaPhi_ifCombinedAlso. Phi_STAfronGLB if isGlb=true. (WRONG!!! should be Phi_STA) Entries 12559 (number of glb muons) Same histogram as GlbMuon_Sta_Phi (No difference whether we plot phi_GLB or eta_STA!!! ???? ) From code is phiEfficiency[1], filled if recoMuon isStandAloneMuon with recoMuon.StandAloneMuon()->phi().

Why StaEta _ifCombinedAlso, StaPhi _ifCombinedAlso are displayed?? I understand might be useful for calculations, but do not need to display them as they are equivalent to GlbMuon _Sta_Eta and GlbMuon _Sta_Phi respectively. If kept it would make sense to have StaEta, StaPhi, StaEta _ifCombinedAlso, StaPhi _ifCombinedAlso contiguously.

  • StaRh_Frac_inGlb. recHits_STAinGLB/recHits_GLB. Entries 12559 (OK, nglb) Range 0,1. Mean 0.75 RMS 0.12
  • StaRh_inGlb_Div_RhAssSta. recHits_STAinGLB/recHits_STAfromGLB. Entries 12559 (OK, nglb) Range 0,1. Mean 0.999 RMS 0.017. Big peak at 1. From code rhAnalysis[0]

Same kind of plots for Tracker: TkMuon _chi2OverDf,TkMuon_eta, TkMuon _p, TkMuon _phi, TkMuon _pt, TkMuon _q, TkMuon _theta, look much better (especially chi2/DF) than for STA...

* TkRh _Frac_inGlb. recHits_TKinGLB/recHits_GLB. Entries 12559. Range 0,1. Mean 0.25, RMS 0.11. Why are there entries with ZERO tk hits???

  • TkRh_inGlb_Div_RhAssTk. recHits_TKinGLB/recHits_TKfromGLB. Entries 12559 (OK, nglb) Range 0,1. Mean 0.96 RMS 0.017. Big peak at 1. Log scale??? But also many entries at zero. WHY???
  • invalidRg_Frac_inTk. Invalid recHits/rechits_GLB. Range 0,1. Mean 0.012, RMS 0.023. Big peak at 0, but also entries at 0.5. No entries at 1... then how can it be no hits from tk in GLB fit?? When is hit not valid???

  • muonReco. Muon reconstructed tracks. Change to Provenance of muon reconstructed tracks. Entries 242979 (all muons???) Maybe log scale to see calo muons??? How to check nentries in each bin??

  • muVstkSytemRotation_negMu (missing s in the hname) pt_TK/pt_GLB vs pt_GLB for mu-. Same for positive muons. Put reference plots and examples with different rotations.

MuonSeed

In instructions. Main histograms to check: Same kinematics variables of the muon ones, which have to be equal one to each other. (looks unclear to me)

Code: /CMSSW/DQMOffline/Muon/src/MuonSeedsAnalyzer.h DQM monitoring source for muon track seeds.

  • NumberOfRecHitsPerSeed. Number of seed recHits (change to number of rechits per seed!) Is zero!!! Explain why or get rid of histo!!! Entries 1196723 (number of seeds???)

What's TSOS??? Are all error histos needed?? Many should go in log scale: seedEtaErr_, seedPhiErr_, seedPt_

SegmentTrackAnalyzer

In instructions. Main histograms to check: Percentual of recHits lost in the track fit, Segment/hits provenance in both tracker/STA regions,

Code: /CMSSW/DQMOffline/Muon/src/SegmentTrackAnalyzer.h: DQM monitoring source for segments associated to the muon track

Both show big peak at 0. Consider log scale? What's the meaning of these plots for SA muons?

Looking at the code CSC/DT TrackHitDivtrackSegmVsEta,Phi,Pt gives the percentual number of CSC/DT hits associated to the track (total number of hits in CSC/DTs div by number of CSC/DT hits associated to track) as function of Eta,Phi,Pt of the track. TrackHitDivtrackSegmHitVsEta,Phi,Pt gives the percentual number of CSC+DT hits associated to the track as function of Eta,Phi,Pt of the track.

I find it quite difficult to understand these plots. Useful for what? What would be ideal situation?

  • trackHitSAProvenance_cosmicMuons(14119 entries) and _globalCosmicMuons (779 entries): Provenance of SA hits in the track. If at least 1 DT hit is used in the track fit and at least 1 RPC hit is also used then the provenance is set for that track as DT+RPC (?) DT flag for SA, glb tracks with hits in DT only (no associated hits in CSC or RPC) Should correspond to different illuminiation of different detector regions (?)
  • trackHitTkrProvenance_cosmicMuons (0 entries) and _globalCosmicMuons (8435 entries) : Number of rec hits from tracker in the SA or glb track.
By definition has no meaning to plot this for cosmicMuons!. This is different from previous (SA provenance) as in here there is one entry per associated hit in any track detector, this is not simply a flag as before.

How to treat this in case one subdetector is absent for some known reason? Different categories of good muons: w/o CSC's, etc... How to treat acceptance holes? MC simulation of experimental conditions?

  • TrackSegments_cosmicMuons and _globalCosmicMuons. Nb of segments associated to the track in DT, DT+CSC or CSC. How this distribution should be? Maximum illumination at 3 segments for DT region? For the rest??

Should we have two separate columns for _cosmicMuons and _globalCosmicMuons

MuonIDDQM

In instructions: Main histograms to check: Residuals between segments and track on x,y and number of matching segments for each DT-CSC, Number of chamber with matching segments and number of total matches.

Code: /CMSSW/DQMOffline/Muon/interface/MuonIdDQM.h.Makes and fills lots of histograms using the various reco::Muon methods. All code is adapted from Validation/MuonIdentification

hSegmentIsAssociatedBool: Segment is Associated Boolean: ? Entries 10202052. I would set range 0,2 Is this for glb,trk,SA? Refers to segments in DTs? When is the plot not OK? No description available.

Looking at the code input collection is "muons". Checks for reconstructed CSC ("cscSegments") and DT ("dt4DSegments") segments whether they are associated or not to any muon object.

GlobalMuons

When are this OK, not OK? Check entries in first histos=integral number of segments? For what are these histos useful? Action to be taken? Description. When number of entries is not OK? Example CSC Station 4 has 1 segment only. Number of entries, means, rms reported for station 1 as example.

  • hCSC1DdXdZ: CSC Station 1 Delta DxDz: Entries 656 (should be number of segments). Range -1,1. Mean ~0.0003, RMS 0.1.
  • hCSC1DdYdZ: CSC Station 1 Delta DyDz: Entries 656. Range -1,1. Mean 0.03?? , RMS 1!!!
  • hCSC1Dx: CSC Station 1 Delta X: Entries 656. Range -50,50. Mean 1, RMS 10
  • hCSC1Dy: CSC Station 1 Delta Y: Entries 656. Range -50,50. Mean -0.3, RMS 12
  • hCSC1NumSegments: CSC Station 1 Number of segments (No arbitration) ?: Entries 832 (number of muon objects? No??) Range 0,10, Mean 1, RMS 1.2
  • CSC Station 1 Pull DxDz: Entries 656. Range -20,20 (so big?) Mean -0.01, RMS 5 (shouldn't it be 0,1?, gaussian fit? how are pulls calculated?)
  • CSC Station 1 Pull DyDz: Entries 656. Range -20,20 (so big?) Mean 0.3, RMS 6
  • CSC Station 1 Pull X : Entries 656. Range -20,20 (so big? -5,5 seems enough) Mean 0.14, RMS 1.0
  • CSC Station 1 Pull Y : Entries 656. Range -20,20 (so big? -10,10 seems enough) Mean -0.01, RMS 1.4

Looking at the code: Fill hCSCNumSegments if "muon" object (glb in this case) crossed or nearly crossed a chamber. This is why it is not simply number of reco muon objects of type glb/trk, and varies for diff stations. To be better understood.

Very few entries for station 4 due to acceptance of cosmic muons.

Same for DTs

  • DT Station 1 Delta DxDz: Entries 9274. Range -1,1
  • DT Station 1 Delta DyDz: Entries 8551. Why different from previous? Range -1,1 Big RMS 0.3 (previous 0.07). Same in every station.
  • DT Station 1 Delta X : Entries 9274. Range -100,100
  • DT Station 1 Delta Y : Entries 8551. Range -150,150
  • DT Station 1 Number of segments (No arbitration) ?: Entries 10917 (?) Range 0,10
  • DT Station 1 Pull DxDz: Entries 9274. Range -20,20 (so big?)
  • DT Station 1 Pull DyDz: Entries 8551. Range -20,20 (so big?)
  • DT Station 1 Pull X: Entries 9274. Range -20,20 (so big?)
  • DT Station 1 Pull Y: Entries 8551. Range -20,20 (so big?)

  • hNumChambers: Number of Chambers. Entries 12559 (number of "muons" "isGlobalMuon"). Peak at 4, mean 3.4 RMS 1.1. Some with 0, some with up to 8.
  • hNumMatches: (?) Number of Matches(?). Entries 12559. Peak at 4, mean 2.8 RMS 1.2. some with 0, some with up to 6

TrackerMuons

Same histos as before but with more entries... There are more tracker muons than global muons, but should they have more segments in the DT,CSC stations??

  • hNumChambers: Has 15933 entries instead of 12559, should be "muons" "isTrackerMuon" . More tracker muons than global muons... No entries for bin 0, why??? Up to 8 chambers
  • hNumMatches : Has 15933 entries instead of 12559, should be "muons" "isTrackerMuon" . More tracker muons than global muons... No entries for bin 0, why??? Up to 6 chambers

Check on muon track quality

PUT ALL THIS IN A SUBFOLDER???

In Instructions: For all the different kind of muon tracks: * TKTracks * cosmicMuons * cosmicMuons1LegBarrelOnly * cosmicMuonsBarrelOnly * cosmicMuonsEndCapsOnly * cosmicMuonsNoDriftBarrelOnly * globalCosmicMuons * standAloneMuonsBarrelOnly * standAloneMuonsEndCapsOnly. Main histograms to check: chi2 / DF, distance of the closest approach and its components, tranverse momentum and its proiection on x,y,z

Code: /CMSSW/DQM/TrackingMonitor/

Tests/MuonRecoTest

In Instructions: Test on track kinematics. Main histograms to check: efficiency on pseudorapidity and radial direction of sta muons
NO REFERENCE PLOTS AVAIBLE, code improvements on going

Code: /CMSSW/DQMOffline/Muon/src/MuonRecoTest.h. DQMOffline Test Client check the reconstructions efficiency of Sta/Glb on eta,phi parameters.

TestSummary

Code: /CMSSW/DQMOffline/Muon/src/MuonTestSummary.h. New tests for muonId parameters

  • energySummaryMap. Energy deposits test summary. For GLB,TK,STA muons checks deposits in ECAL,HAD abd HO (NA now). TK and STA must not be GLB in this case. RED for HAD TK. Based on ecalS9PointingMuDepositedEnergy and hadS9PointingMuDepositedEnergy. Fit to a langau with a set of initial values in range 0.04,3 and 0,7. Checks hpeak between 0.2 and 0.3 and between 2 and 3.5 respectively. Is this really significant test? Does it check that there are enough events to perform tests???Looking at the original histo few events (131). Does it check that fit has properly converged??? (I guess his is if(fit) in the code). Tests on errors of final parameters???Why is not hadS9PointingMuDepositedEnergy_Tk_muons shown in red in Muon/MuonEnergyDepositAnalyzer??? Looks to me peak for HAD more between 1.5 and 3 than between 2 and 3.5. Theory??? Energy calibration???

-- SilviaGoyLopez - 12 Aug 2009

Edit | Attach | Watch | Print version | History: r13 < r12 < r11 < r10 < r9 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r13 - 2009-08-19 - SilviaGoyLopez
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Sandbox All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback