(X)DIGI2CSV#
The (X)DIGI2CSV repository is a versatile tool for running Allen and Moore algorithms in a reproducible manner. While its primary function is to convert (X)DIGI files on the grid into CSV or PARQUET formats, it can also be used for a variety of other tasks. For example, it can convert (X)DIGI files into MDF or ROOT, and MDF files into CSV. The repository also makes it easy to process large sets of files in parallel by quickly submitting jobs to HTCONDOR.
Every program can be configured using a YAML file such as the one below, which can be run simply by using ./run/run.py -c the_configuration_file.yaml.
moore_input:
nb_files: 3
start_index: 0
# Input file is in the grid
bookkeeping_path: "/MC/Upgrade/Beam7000GeV-Upgrade-MagDown-Nu7.6-25ns-Pythia8/Sim10aU1/30000000/XDIGI"
dddb_tag: "dddb-20210617"
conddb_tag: "sim-20210617-vc-md100"
evt_max: -1
xdigi2csv:
detectors: # detector hits or clusters to dump
- velo
- ut
- scifi
# Extension and compression
format: parquet
compression: lz4
output:
outdir: "minbias-sim10aU1" # output directory of the CSV files
computing:
program: xdigi2csv # program to run
This allows you to define jobs in a way that is quick, easy, and reproducible. By using the YAML configuration file, you can easily configure the program to suit your needs, whether you need to convert (X)DIGI files, generate MDF or ROOT files, or process large sets of files in parallel.
(X)DIGI2CSV is being developed as part of the ANN4EUROPE project, with the goal of training neural networks for tracking at LHCb.
Danger
Please note that the (X)DIGI2CSV repository was developed primarily for internal use and may have limitations. We do not recommend its use by external parties.