PID correction of simulation

Introduction

The following set of scripts aims to do two main things:

• Create the histograms that parametrize the efficiency of specific PID cuts for muons, kaons and pions both in Data and MC
• Attach these efficiencies as weights to the MC samples, to then use their ratio to correct the simulation.

Creating the histograms

A preliminary version of the histograms can be produce by running:

create_pid_maps -b nobrem -p Pi -s b1 -o /path/to/maps -r signal -m 1

where:

options:
  -h, --help            show this help message and exit
  -b {nobrem,brem}, --brem {nobrem,brem}
                        Version of binning file
  -p {e,Pi,K,Mu,P}, --particle {e,Pi,K,Mu,P}
                        Particle name
  -s SAMPLE, --sample SAMPLE
                        Sample/block, e.g. b1, b2...
  -o OUT_DIR, --out_dir OUT_DIR
                        Directory where pkl files will go
  -r {signal,control}, --region {signal,control}
                        Used to define selection
  -d, --dry-run         Enable dry-run mode (default: False)
  -m MAXFILES, --maxfiles MAXFILES
                        Limit number of files to this value
  -v, --verbose         Will print debug messages

the config and binning files are in:

rx_pid_data/config/config.yaml
rx_pid_data/config/binning.yaml

in versioned yaml and json files respectively. In the config files the samples are nicknamed as:

b1  : 2024_WithUT_block1_v2
b2  : 2024_WithUT_block2_v2
b3  : 2024_WithUT_block3_v2
b5  : 2024_WithUT_block5_v2
b6  : 2024_WithUT_block6_v2
b7  : 2024_WithUT_block7_v2
b8  : 2024_WithUT_block8_v2

and the b* key is what the -s flag should get.

Cluster jobs

In order to make these maps with jobs sent to the cluster do:

• Move to LXPLUS
• Copy to an empty directory both submission and running scripts in src/rx_pid_data/jobs and src/rx_pid_data_scripts respectively.

chmod +x make_pid_maps

# Modify the line OPATH to point to the user's directory where the outputs will go

# The log files will be placed here, the directory has to exist or else the jobs will be held
mkdir logs/

condor_submit job.sub

the jobs can be monitored with watch condor_q

Plotting

To make plots do:

plot_histograms -d /directory/where/pickle/files/are

and the utility will loop over all the pkl files plotting the histograms inside.