Software wishlist

General

  • Provide QA macros from the various analyses of the PWGs; short how-to documentation needed --> high priority --> QA workshop (CCs, convenors, Aaron, Jenny, Jenny, EMP?)?
  • Update/extend tutorial macros, essential as documentation; what, who (Klaus G.?) --> high priority --> follow-up with Klaus G.

Electromagnetic processes (Alaa, Anna, Stefan)

  • An interface at the level of event generator that allows the users to weight the generated PHSP events with some functions (cross section). This is important for the three ongoing analyses (GPD, unphysical region and GDAs) where currently no dedicated event generators for the corresponding processes are available. JM: Use weight factor and random number generator to decide to keep event ("hit and miss"). Tobias: write own event generator using filter? Priority low.
  • Event filtering needed to generate high amount of background in the specific region (GPDs, Stefan). The existing filters were working for older versions but not with the new version on the cluster. This problem seems to be solved and it was due to some configuration problems. The existing filters are sufficient for the current analysis purpose. However on the long term it would be nice if there would be an easy way to define own filter functions on the generators like for t and Q2 (in GPD analysis) etc. within the user interface (currently one has to change the code and recompile things); Define own functions in a user interface? Priority low, long term goal.
  • Improvement of the Muon system PID. The current version of the PID algorithm is based on a very simplified parameterization. This is important to separate muon from the background pions. People are developing their own selection code for example based on TMVA (form factors and now GPDs) however it would be good to have a common method implemented in PANDAROOT that can be easily used by all users, in particular that some times a difference in the results of different analyses is observed. Tobias: long standing problem. Principle working for the moment... Priority low. ---> check classifier of EMC, very similar!!!
  • Updates on the PANDARoot releases: It is not easy to find directly the reason of the differences between the results using new PANDARoot releases (due to some improvements in the code or due to new bugs ). It would be very helpful to have a short documentation (or a talk in the CM) about the release summarizing the main modification and also including few plots about the main features for example on the reconstruction efficiencies of charged tracks and PID efficiencies, which could be used as a reference for the individual analyses. (maybe these info already exist but I missed them. Backtracing of problems sometimes difficult; Tobias: changelog, developers supposed to write information in the log; QA task checks features; Get analysis macros from PWG --> QA task; quality assurance for P2 paper; weekly/monthly basis.
  • Concerning PandaRoot features - it would be very helpful (especially for big data of backgrounds) to have an option to run the scripts with use of parallel technologies (if it is possible in principal). --> event-based parallelization is present, long term paralelization in detail... need for online reconstruction, not so much now offline --> low priority
  • As for PID algorithms for Muon system - we (Dubna muon system group) are working on generalization of experimental data (from muon prototype) for pion to muon separation, which are planned to be used in the future for simulation. But due to the lack of the man power it will take quite a long time.. Not earlier than in two years!

Charmonium-like and exotics (Frank and Marc)

  • Tracking of charged particles at small angles --> within acceptance of forward spectrometer (Aaron, J/psi studies); Tobias: ongoing, work in process group Jagelonian group; Proposal: look at efficiency of pT distributions. Might be limit of detector itself, not software; skewed layers start at middle? Action: tobias+analysists produces pT distributions, efficiency! Email to Tobias to exchange plots from recent studies of Aaron.
  • Pattern recognition for secondary tracks as well --> new version in development branch from Anna; once pZ issue are fixed and curly tracks --> release!
  • Implementation of pair production (gamma->e+e-) recovery ---> Action: Aaron reported, check with Aaron what it all means? --> useful to integrate in PandaROOT?
  • Gap in the detection efficiency, likely due to problem of GEM geometry implementation --> solved in development branch
  • Event filter also working on new Lustre --> fixed
  • MC truth matching within the decay tree fitter, no implementation for neutrals --> Ralf business, neutrals are bigger issue, need for covariance matrices; action: contact with aaron, since he is doing something... HIGH priority
  • Low efficiency in J/psi->e+e- channel, likely due to PID issue --> new PID implementation might be the cause, still looking into this.
  • EMC shower profile corrections for cluster with missing crystals --> EMC group task, edge correction being done, will go into PandaRoot.
  • EMC updates, next release, important for PWG. ~mid 2022, earliest for release. Proposal to use develop branch. PWG wish: full reconstruction of neutral, covariance matrix kin. fit.

Hyperons (Michael, Josef/Marcel)

  • Realistic tracking in barrel+forward regions, resolve discrepancies between analysis groups -> work in progress, likely next release; already in development branch from Anna!
  • Containerised version or virtual machine of pandaroot to ease the installation procedure --> available!!! docker container available + documentation.
  • Realistic PID welcome, but less critical for the hyperon studies, more important for Phase Two studies --> unclear when this is fixed
  • Residual target distribution including both transversal and longitudinal distributions, low priory though --> Ralf is working on this.

Overview

Item Priority Description Contact person Progress Comment

-- JohanMesschendorp - 19 Nov 2021
Topic revision: r6 - 2022-01-24, JohanMesschendorp
Copyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Panda Wiki Send feedback | Imprint | Privacy Policy (in German)