pidcalib : The efficient processing of LHCb/PidCalib PID calibration data using ostap project
- mandatory: ostap
- The project allows to get 1D, 2D or 3D efficiency historgams for particle identification using the calibraton samples
- Calibration samples can be accessed either from CERN eos or via GRID
- the first way requires read access to
/eos/lhcb/grid
partition - the second way require to have valid Grid proxy and access to LHCbDirac machinery via cernvmfs
- both conditions are satisfied on CERN machines,
lxplus.cern.ch
in particular
- the first way requires read access to
- The calbration samples can be processed in parallel using pathos-based parallelisation
- The output histograms and useful processing information are placed in the outptu database, default is
PIDCALIB.db
To use this function a helper script needs to be created, see example in examples directory. The script to be execute from command-line
pid_calib2.py
All command-line arguments can be inspected using -h
flag
ex_pidcalib_run2.py -h
- If calibration samples to be taken from the Grid, one needs to ensure that scripts in the directory scripts are in the path, e.g.
PATH=$PATH:<...>/scripts ex_pidcalib_run2.py ...
Also for this mode one needs
-
access to LHCbDirac at
cernvmfs
, see the content ofdirac-command
script in scripts directory -
valid Grid proxy
-
In case
/eos/lhcb/grid
is accessible directly, no Grid proxy and machinery is requred. In this mode, a powerfull pathos-based parallelization is available, activated with-z/--parallel
keys
ex_pidcalib_run2.py --parallel
- If
/eos/lhcb/grid
is accessible directly, but input data are requested from the Grid, the optional conversion using the comman line flag--useeos
is possible. It opens a way for the parallel processing.
ex_pidcalib_run11.py P -s 20 -p MagUp -c 'P_hasRich==1'