Skip to content
Snippets Groups Projects

Mcbm kfparticle mr

Closed Yue Hang Leung requested to merge y.h.leung/cbmroot:mcbm-kfparticle-mr into master

Merge request reports

Merge request pipeline #31362 passed

Merge request pipeline passed for cadf24cf

Closed by Sergey GorbunovSergey Gorbunov 1 month ago (Jul 3, 2025 5:25pm UTC)

Loading

Activity

Filter activity
  • Approvals
  • Assignees & reviewers
  • Comments (from bots)
  • Comments (from users)
  • Commits & branches
  • Edits
  • Labels
  • Lock status
  • Mentions
  • Merge request status
  • Tracking
  • @y.h.leung,

    • all the automatically generated files should be removed from the merge request (e.g., gphys.dat, TRhistos.root), as well as backup files (e.g., .run_reco_digievent.C.swp)
    • what is the difference between the /macro/beamtime/mcbm2022/mcbm_reco_kf_*.C files? As I see, there are different versions of them. Do the macros have different purposes? If no, please select the latest working one and remove the others from the merge request
  • I think we would need to do first some local squashing to simplify the commit history (remove duplicated commits, remove commit reverting each other with many files, ...) and reduce the number of changed files.

    Should we sit together at some point tomorrow to do it?

  • Hi Pierre, Norbert will take me to GSI tomorrow morning, and that will be very helpful. Thanks!

  • added 3 commits

    Compare with previous version

  • added 1 commit

    • 49c8caab - Update KFParticleFinder for mCBM Lambda analysis

    Compare with previous version

  • requested review from @se.gorbunov

  • added 1 commit

    • cfd326b1 - fix to pass CI test + add setter to kfpartcleanalysis

    Compare with previous version

  • Yue Hang Leung added 3 commits

    added 3 commits

    • 05191d25 - 1 commit from branch computing:master
    • a627ce9b - Update KFParticleFinder for mCBM Lambda analysis
    • 27836d86 - fix to pass CI test + add setter to kfpartcleanalysis

    Compare with previous version

  • added 1 commit

    • cadf24cf - fix to pass CI test + add setter to kfpartcleanalysis

    Compare with previous version

  • Dear @f.uhlig, @v.friese, @p.-a.loizeau, @se.gorbunov, @s.zharko,

    you have been identified as code owner of at least one file which was changed with this merge request.

    Please check the changes and approve them or request changes.

  • Pierre-Alain Loizeau resolved all threads

    resolved all threads

  • mentioned in merge request !1942 (merged)

  • @s.zharko Hi Sergei,

    Here is a summary of what the codes are doing: Macro: macro/beamtime/mcbm2024/mcbm_reco_kf_lambda.C This macro takes data digivents as input, performs hit reconstruction, track reconstruction, and KFParticle finder. KFParticle finder is called on L604

    The first task is kfParticleFinderPID The macro uses: kfParticleFinderPID->UseTopoPID(); The code is located at reco/KF/CbmKFParticleFinderPID.cxx L663 It calculates the DCA defined by two STS hits L667-708, and pushes back the PID on L727

    The next task is CbmKFParticleFinder kfParticleFinder->SetmCBMmode(); sets fmCBM to true. kfParticleFinder->UseOrigin(); uses 0,0,0 as vertex The code is located at reco/KF/CbmKFParticleFinder.cxx At L195, because fmCBM is true, global tracks are used instead of STS tracks From L233-273, the T0 is found by looping over TOF hits. Then the time for each TOF hits is reset by subtracting T0 Some cuts are skipped when using mCBM mode. Because the tracking is still not finalized, there is no need to set such cuts as default The next part is on L358-425, where velocity is calculated according to the time and the length. Then, using the PID, q/p is calculated, and added to the track parameters. The covariance matrix is also updated here L493-494. We skip events with less than 1 proton and 1 pion because you cannot reconstruct lambda from these events L554: Set vertex to 0,0,0 L571-588: The B-field is 0, so we just set to 0. The vChiToPrimVtx is set to 999. This value ensures that for mCBM all tracks will enter reconstruction. There is no proper primary vertex reconstruction so we just set this to a large value

    The next task is CbmKFParticleFinderAnalysis This task is to save the lambda candidate information. The lambda_tree is a ttree and is defined in L98-149 L278 is the beginning of the loop for kfparticle. It basically loops over all particles, and when the PDG code is 3122 (lambda), it saves the information, including pT, mass, decay length, etc.

    Let me know if you need more information.

    Best, Yue Hang

  • added mCBM label

  • Loading
  • Loading
  • Loading
  • Loading
  • Loading
  • Loading
  • Loading
  • Loading
  • Loading
  • Loading
  • Please register or sign in to reply
    Loading