v01-36-01

v01-36-01

• 2024-09-09 jmcarcell (PR#139)

  • Add a missing #include <algorithm>

• 2024-09-03 Thomas Madlener (PR#137)

  • Remove CentOS7 from the Key4hep CI workflows

v01-36

v01-36

• 2024-07-30 Ulrich Einhaus (PR#136)
  Bug: PFOs were ignored if their MC PDG was not among signal or background PDGs. This is of minor effect, since by default all detector-stable charged particles are considered signal or background, but could lead to MC info leaking into reconstructed values in case of unintended usage.
  Solution: This effect now requires training mode to be ON, which is exclusive with inference mode.

• 2024-06-24 tmadlener (PR#135)

  • Add a ReconstructedParticleParticleIDFilterProcessor that allows to filter ParticleID objects from existing ReconstructedParticles.

• 2024-06-24 tmadlener (PR#132)

  • Make the TrueJet processor use the PIDHandler to set the ParticleIDs for the different objects it creates. This sets the necessary metadata that is required, e.g. for the conversion to EDM4hep.

• 2024-06-19 Carl Mikael Berggren (PR#134)
  Reduce the size of the ParticleID vector for the final fermion-antifermion pair, since

  for this case, there can only be one pair. This to avoid cluttering of empty collections after transition to the EDM4HEP world. At the same time, the documentation and example steerings in the examples subdirectory have been updated. mainly for the move of TrueJet_Parser from here to MarlinUtil, but also spell-checking etc.

• 2024-06-10 Ulrich Einhaus (PR#133)

  • This adds the WWCategorisationProcessor to MarlinReco
  • It categorises each event by its WW decays channels. It provides a true category (only meaningful for true WW events) as well as two levels of reconstructed category. They are stored as event parameters.
  • This may serve as common coherent categorisation for any analyses using WW events.

• 2024-05-07 Bohdan Dudar (PR#99)

  • Fix all compiler warnings in MarlinReco, including
    • A lot of shadowed variables
    • A lot of unused parameters / variables
    • A few deprecations
    • A genuine use-after-free bug
    • A few others
  • Make at least one CI workflow use -Werror to make it harder to (re-)introduce new warnings

• 2024-04-16 NAKAJIMA Jurina (PR#131)

  • Fixed PDG code for kinks identifies as antiSigma+

v01-35

v01-35

• 2024-02-23 Ulrich Einhaus (PR#130)
  Minor changes in CPID:

  • allow for no automatic creation of root file or plots
  • if no output root file and no training then no intermediate TTree is filled
  • add efficiency numbers on additional confusion matrix plot
  • in TOF algorithm use mass² instead of beta

  LeptonID: Remove inheritance from EventModifier

• 2024-02-16 tmadlener (PR#129)

  • Populate the type information for the output relation collection in the SimpleFCalDigi
  • Fix some warnings there as well. (Fixes #100)

• 2024-02-12 tmadlener (PR#128)

  • Use the key4hep-build action for building against key4hep stacks
  • Switch to LCG_104 based clicdp nightlies for CI

• 2024-02-12 tmadlener (PR#126)

  • Make the KinkFinder always produce all output collections even if some of them are empty

v01-34

v01-34

• 2023-07-07 tmadlener (PR#124)

  • Usage of new utility functionality requires a newer version of MarlinUtil

• 2023-07-07 yradkhorrami (PR#123)

  • Merge SLDCorrection to MarlinReco
  • First implementation of SLDCorrection: for the time being, done only on semi-leptonic decays of bottom hadrons in b-jets

• 2023-07-07 Leonhard Reichenbach (PR#121)

  • LeptonID: remove notebook output

• 2023-07-04 Ulrich Einhaus (PR#118)
  Comprehensive Particle Identification Processor: First beta version of CPID.
  Extracts PID-related observables from PFOs and combines them in a training model, which can be inferred afterwards to data.
  This release contains the initial version of the Marlin processor with an example steering file, the input algorithm generic class + manager along with a library of different input algorithms, corresponding to different PID observables, as well as the training model generic class + manager along with a library of predefined models for training and inference.
  The selection of algorithms and models is handled via the processor steering file.
  The managers allow to add new modules (algorithms and models), compile and use them via the steering file, without the necessity to touch any other files.
  ComprehensivePIDProcessor.h contains a ReadMe section with an explanation of all processor steering parameters.

• 2023-07-03 Bohdan Dudar (PR#117)

  • In case of the option to measure time-of-flight at the SET: now using raw (digitized strips) SET hits instead of simulated hits. It doesn't change anything now, but could become trouble in the future when proper time digitization is added.

• 2023-06-29 Leonhard Reichenbach (PR#120)

  • LeptonID: update weights

• 2023-06-13 tmadlener (PR#115)

  • Bump the minimal required cmake version to 3.12
  • Fix the order in which DD4hep and LCIO appear in the libraries to link against to make sure to not pick an inappropriate version from the underlying environment

• 2023-06-13 Leonhard Reichenbach (PR#114)

  • Added LeptonIDProcessor to identify electrons and muons in jets using boosted decision trees.

• 2023-06-12 tmadlener (PR#116)

  • Switch from dd4hep::long64 to dd4hep::CellID to be compatible with DD4hep after AIDASoft/DD4hep#1125

• 2023-06-07 Bohdan Dudar (PR#113)

  • Always use the first (before) --> latest (now) curl in the track to get extrapolated track position at the calorimeter surface. This gives sometimes better estimate of the track position at the ECAL surface, especially for the tracks with large number of curls. Thus new version gives better time of flight correction for the distance to the surface and thus TOF itself.
  • Minor style improvements

• 2023-06-07 Bohdan Dudar (PR#112)

  • Switch to the helix formula without Omega: $\ell_{i} = \frac{|z_{i+1} - z_{i}|}{|\tan{\lambda_{i}}|}\sqrt{1 +\tan^2{\lambda_{i}}}$. It shows the best performance so far.
  • Other bug fixes and consistency improvements.

• 2023-05-11 Julie Munch Torndal (PR#110)

  • Added CheatedMCOverlayRemoval processor to identify MC particles that are overlay and remove the corresponding PFOs from the collection.
    • See under Analysis/OverlayRemoval/example/CheatedMCOverlayRemoval.xml for how to run processor

• 2023-02-07 Bohdan Dudar (PR#108)

  • Make encoding RecoParticle relation weights more explicit with a new encode function in MarlinUtil

v01-33-01

v01-33-01

• 2022-12-06 Thomas Madlener (PR#107)

  • Remove no longer available CI workflow based on gcc8 since the underlying nightly builds are no longer available
  • Update github actions to latest available versions

• 2022-12-06 Bohdan Dudar (PR#106)

  • Fix a seg. fault, in rare cases, when the track fit fails in both directions due to the lack of hits.

v01-33

v01-33

• 2022-04-19 Bohdan Dudar (PR#104)
  • Migrated track length and mean harmonic momentum code from TOFEstimators into separate TrackLengthProcessor to save CPU computing time.
  • NOTE: If you have been using these processors outside of the standard reconstruction chain you will have to update your steering files (see iLCSoft/ILDConfig#133 for the necessary changes)

v01-32-01

v01-32-01

• 2022-03-18 Bohdan Dudar (PR#103)
  • Try to fit subtrack forward if backward fit fails
  • Fix seg. fault if the very first sub track fit fails in both directions
  • Change behaviour of checking track direction. Now it is more robust and relies on z position of first and last hits.

v01-32

v01-32

• 2021-11-05 Bohdan Dudar (PR#96)

  • Major upgrade of TOFEstimators with bug fixes and updates that are not backward compatible
    • Track length calculation is significantly improved by refitting the track and iterating over the track states on each tracker hit
    • Fixed bugs with phi coordinate flip and wrong abs() instead of std::abs()
    • Output in the PIDHandler is changed to the minimal set of required three parameters for the PID (momentum, track length, TOF)
    • Input steering parameters adjusted
    • Removed TOFPlots processor. General clean up of the files and the code
    • Detailed documentation is added

• 2021-11-04 Thomas Madlener (PR#101)

  • Fix all warnings that are trivial to fix (i.e. where the fix is obvious without having to think about things). .
    • Make all Processors that have pointer members have deleted copy c'tors and assignment operators.
  • Fix a potential out ouf bounds access in TrueJet.

  Thanks to Bohdan Dudar (@dudarboh)

• 2021-11-01 Thomas Madlener (PR#94)

  • Make the IsolatedLeptonTagging processor always produce the expected output collections, even for empty input collections, i.e. if the inputs are empty:
    • the OutputPFOsWithoutIsoLepCollection will simply have the same content as the InputPandoraPFOsCollection
    • the OutputIsoLeptonsCollection will be empty.
  • On the other hand actually missing input collections will now no longer be handled as these point to a real problem (e.g. typo in the collection name). Fixes #93

• 2021-09-29 Frank Gaede (PR#98)

  • apply the correct units for TPC parameters from DD4hep
    • fixes #97
  • apply correct units for VXD and SIT layers
    • need cross check of efficiency for kink finding
    • (how could this have worked before ?)

• 2021-08-25 Andre Sailer (PR#95)

  • CI: build against LCG_99python2 gcc8 and LCG_100 gcc10, clang11

v01-31

v01-31

• 2021-06-15 Thomas Madlener (PR#92)

  • Move the TrueJet_Parser utility class to MarlinUtil
    • Make it possible to use it outside of MarlinReco
    • Keep example usage and README here in MarlinReco for better discoverability

• 2021-05-07 A. Irles (PR#91)

  • PIDTools:
    - LikelihoodPID modified to be able to accept as arguments the name and type of method to run and store (allowing for versioning)
    - AngularCorrection_dEdxProcessor created to use with DST samples generated with v02-02 and v02-02-01. The dEdx is corrected for angular effects.
    - Compute_dEdxProcessor2021, modifictation of the default Compute_dEdxProcessor
    - The angular correction is optional
    - The function of this correction is a pol3(lambda)
    - It is applied AFTER smearing
    • Example steering files are added.

v01-30

v01-30

• 2021-03-03 Remi Ete (PR#83)
  • RealisticCaloDigi: Added new option for energy/charge and time integration
    • deal with slow/fast shaper of ROC chips and time estimate
    • new processor parameters:
      • integrationMethod: "Standard" for old implementation (default) or "ROC" emulating the behavior of ROC chip.
      • fastShaper: fast shaper time, unit in ns. Only for ROC method
      • slowShaper: slow shaper time, unit in ns. Only for ROC method
      • timingResolution: optional time resolution gaussian smearing to apply, unit . Only apply is > 0. Default is 0 (no smearing)