Parametric simulation

From Charm-Tau Detector
Revision as of 11:02, 15 April 2020 by Razuvaev (Talk | contribs)

Jump to: navigation, search

The new parametric simulation package is described here.

Contents

Talks


SctParSim (Current)

How to run

Login to stark or proxima machine.

ssh stark -X
setupSCTAU; asetup SCTauSim,master,latest
mkdir workarea
cd workarea
mkdir run
cd run
cp /home/razuvaev/public/misc/pi_ms_f1_mppc2_px3_d200_mla4_graph2d.root .
cp /home/razuvaev/public/misc/gun1.cfg .
cp /home/whitem/public/misc/pi_p.root .
cp /home/whitem/public/misc/pi_m.root .
cp /home/whitem/public/misc/mu_p.root .
cp /home/whitem/public/misc/mu_m.root .
runparsim.py

Options

Option Run example Description
-b, --batch runparsim.py -b Turn on the batch mode. Suppress EventDisplay execution.
-c, --change runparsim.py -c my_cfg_file.dat Change some parameters of the detector.
-n, --neve runparsim.py -n 31415 The number of events to process.
-g, --gun runparsim.py -g Turn on the particle gun mode.
-ig, --input-gun runparsim.py -g -ig my_gun.dat Input particle gun configuration file.
-i, --input runparsim.py -i gen_dkpipi0.root Specify the input file.
-o, --output runparsim.py -o parsim_dkpipi0.root Specify the output file.
--profile runparsim.py --profile Turn on the cProfile to analyse the performance.

Event display

The event display has two projections: x-y and y-z. All detector subsystems are presented. Some of them overlay, especially PIDs, it is not important for parametric simulation, but let's study several options at once. All particles from AllGenParticles branch are presented at the plot by lines of different styles, colours and thicknesses. Somehow the line thickness corresponds to the particle mass, the wide line --- the more massive particle. Warm colours are devoted to positive charged particles and cold to negative ones. Also, particle lines are labeled.

The event display is switched on by default. To switch it off run simulation with -b option.


Detector configuration

The detector parameters can be changed via a configuration file my_cfg_file.dat placed in the main simulation folder. The file has a simple structure --- one parameter and its value(s) per line. A parameter's name and value should be separated by spaces. Arrays should be written in [] brackets. The values in arrays should be separated by commas. Empty lines and lines contained incorrectly parameter names are ignored.

In the example below the parameter at the first line and the second line is one number, while the parameter at the second line is an array.

trck.minPt 0.05

trck.corrMtx.pij -0.08

calo.resPar [0.167,0.0,0.011]

The parameters can be given in any order.

All detector parameters can be viewed in the configuration files:

Subsystem Congiguration file Name to change parameters
ASHIPH ashiphpars_std01.json ashiph1030
Calorimeter calopars_std01.json calo
FARICH farichpars_std01.json farich
FDIRC fdircpars_std01.json fdirc
Muon system muonpars_std01.json muon
ToF tofpars_std01.json tof
ToP toppars_std01.json top
Tracker trkpars_std01.json tracker

The detector subsystems sizes are stored in detlayout_std01.json. They can be changed in the same way.

Particle gun

Examples

Pictures

PAPAS (Old)

About papas, heppy et cetra

Particle propagation is done by geometry calculation. To valid the calculation several different cases were plotted.

Helix.
Helix.

Configure detector parameters

The detector parameters can be changed via a configuration file CTauPapas.cfg placed in the main papas simulation folder. The file has a simple structure --- one parameter and its value(s) per line. A parameter's name and value(s) should be separated by spaces. Empty lines and lines beginning with # are ignored.

In the example below the parameter at the first line is one number, while the parameter at the second line is an array.

ecal_emin_barrel 0.05

ecal_eres 1.34e-2 0.066e-2 0 0.82e-2

The parameters can be given in any order.


Configure detector parameters

The file ctau_input_sim.txt contains two lines. The first line is the path to a primary simulation file (see MC Data Sets page). The second line is an integer number of events to be processed.


How to run papas

Copy a directory with papas on stark the machine and go to this directory.

cd

cp -rf ~razuvaev/myheppy .

cd myheppy

There are a directory output for output files, detector configuration file CTauPapas.cfg, file ctau_input_sim.txt with a path to the file with primary generator events, and the folder heppy with heppy code itself. Let's go into it and tune environment.

cd heppy

source init.sh

Now it is time to run papas. You may be asked a question because the output directory is not empty. So just input y or clean the folder.

cd test

./heppy_loop.py ../../output/ ctau_cfg1.py

If it don't want to run try source ~razuvaev/.bashrc and source ../init.sh because it can be caused by the problem with environment variables.

When papas simulation has been done one need to present papas output to a suitable form and also add initial generator information.

cd ../../

./txt2tree.py

The output root tree is available in the file myheppy/output/txt2tree.root.

Output tree

The output tree contains branches which can be divided in several groups:

  • reconstructed particle parameters;
  • generated particle parameters;
  • generated vertices;
  • connection between reconstructed particles, generated particles and generated vertices.

The table below presents branches and description of their content.

Name Type Length Description
Reconstructed particles
n int 1 The number of reconstructed particles.
px float [] n The reconstructed particle momentum: x coordinate.
py float [] n The reconstructed particle momentum: y coordinate.
pz float [] n The reconstructed particle momentum: z coordinate.
Generated particles
n0 int 1 The number of generated particles.
px0 float [] n0 The generated particle momentum: x coordinate.
py0 float [] n0 The generated particle momentum: y coordinate.
pz0 float [] n0 The generated particle momentum: z coordinate.
Generated vertices
nv0 int 1 The number of generated vertices.
vx0 float [] nv0 The generated vertex: x coordinate.
vy0 float [] nv0 The generated vertex: y coordinate.
vz0 float [] nv0 The generated vertex: z coordinate.
Links
recgen int [] n Transform a reconstructed particle index to the generated particle index.
genver int [] n0 Transform a generated particle index to the generated vertex index.

Analysis example

Here a short analysis example of D^0 \to K_S^0 \pi^+ \pi^- is presented. The things are performed with PyROOT.

The data a taken from the available exclusive sample.

The code can be taken from github [1] or find at the stark cluster: /home/razuvaev/myheppy/search_dkspipi.py.

Personal tools