Warning, /analysis/HF-Particle/QM25_productions/README.md is written in an unsupported language. File is not indexed.
0001 # Running K-short reconstructions on real data
0002
0003 ## Overview
0004
0005 These are the full scripts used for producing various resonance nTuples used for Quark Matter 25. It also contains K-short and D0 invariant mass fitters.
0006
0007 By default, only the K-short reconstruction is enabled but there are several other reconstructions that can be performed. Be aware that they will have significantly more background than the K-short and hence require more events processed to see anything. They are mainly used as examples to write your own reconstruction around.
0008
0009 ## Test 1, running the macro
0010
0011 ```Fun4All_HF_QM25.C``` is a top level macro that takes a **single** track DST and runs reconstructions with selections defined in ```HF_selections_QM25.C```. It makes output files, finds associated trigger and cluster DSTs for the track DST you read and handles all the file movements. It is set to run over 100 trigger frames and find K-short candidates. Run it with
0012
0013 ```c++
0014 root -l -q -b Fun4All_HF_QM25.C
0015 ```
0016
0017 As Fun4All processes, you will start to see information printed to the screen. As long as KFParticle's verbosity is > 0, it will print out each candidates information. You will get the particle masses, momentum and positions amongst other things.
0018
0019 After it finishes, you should have a new folder called ```pipi_reco``` with a file ```outputKFParticle_pipi_reco_00053877_00000_00000.root``` containing the reconstructed candidates. Open this file in the root prompt then type
0020 ```
0021 DecayTree->Draw("K_S0_mass")
0022 ```
0023 to see the pipi invariant mass. There are several output branches in this file with information on the mother, daughters (including detector information like number of TPC states used by ACTS), global information such as the PV position or number of tracks associated to the PV, and trigger information
0024
0025 ## Test 2, running with shell scripts
0026
0027 I normally find it easier to run KFParticle using shell scripts. Run the previous reconstruction using the command:
0028
0029 ```bash
0030 ./runHFreco.sh 10000 DST_TRKR_TRACKS_run2pp_ana475_2024p017_v001-00053877-00000.root /sphenix/lustre01/sphnxpro/production/run2pp/physics/ana475_2024p017_v001/DST_TRKR_TRACKS/run_00053800_00053900/dst/ 0
0031 ```
0032
0033 The same track DST will be analysed, but this time we will run over 10000 trigger frames
0034
0035 ## Test 3, fitting
0036
0037 After Fun4All has finished, the file should now contain 302 candidates. You can fit the selected candidates uses a RooFit macro in the folder ```K_S0```. Run it with
0038 ```bash
0039 root -l -q -b K_S0_Fitter.C
0040 ```
0041
0042 You should now have 3 files for the fit: a root macro, a PDF and a PNG. Feel free to use this fitter to develop further fitting codes.
0043
0044 ## Further things that can be done
0045
0046 All previous instructions should be useful in getting people familiar with KFParticle and simple reconstructions that don't require much CPU resources. It is possible to run multiple reconstructions simultaneously in Fun4All, and to run over lots of data using condor. I won't go into much detail here to avoid oversaturating our condor resources but you can build a condor job list by using a shell script provided here
0047 ```bash
0048 sh getFiles.sh 53877
0049 ```
0050 which will get all track DSTs for run 53877.
0051
0052 You can get as many condor lists as needed with this script. Then open ```submitCondorJobs.sh``` and alter ```runs``` to contain the runs you with to iterate over. After this, submit condor jobs with
0053 ```bash
0054 sh submitCondorJobs.sh
0055 ```
