Skip to content
Snippets Groups Projects
Select Git revision
  • badef9305b1ec1b4e62d6d3e924f8e38a69bca49
  • master default protected
  • beamline_training
  • npsim
  • code.jlab.org
  • pr/nhcal
  • pr/rerun_fix
  • pr/minor_fixes
  • temp/df
  • Simple-Shyam-patch-1
  • interruptible-resource-groups
  • femc-collect-always
  • add_lowq2_benchmarks
  • pr/insert_neutron_flaky
  • Low-Q2_onnx_training
  • Low-Q2_training
  • matplotlibrc
  • pr/normalized_output_branch_sizes
  • snakemake-batch
  • pr/snakefmt
  • pr/venv
  • v0.0.1
22 results

detector_benchmarks

  • Clone with SSH
  • Clone with HTTPS
  • user avatar
    Dmitry Kalinkin authored and GitHub committed
    badef930
    History

    EIC Detector Benchmarks

    pipeline status

    Overview

    Detector benchmarks are meant to test for regressions in individual detector subsystems. The analysis is meant to avoid a reconstruction step. So this precludes using juggler for processing the events.

    Documentation

    See common_bench.

    Adding new benchmarks

    To get an idea of what to do look at an existing benchmark in the benchmarks directory.

    Running Locally

    Local development example

    Here we setup to use our local build of the juggler library. Note juggler is not needed for detector_benchmarks because it is not used but this is the same setup for reconstruction_benchmarks and physics_benchmarks.

    First set some environment variables.

    export DETECTOR=epic             # epic is the default
    git clone https://eicweb.phy.anl.gov/EIC/benchmarks/detector_benchmarks.git && cd detector_benchmarks
    git clone https://eicweb.phy.anl.gov/EIC/benchmarks/common_bench.git setup
    source setup/bin/env.sh && ./setup/bin/install_common.sh
    source .local/bin/env.sh && build_detector.sh
    mkdir_local_data_link sim_output
    mkdir -p results
    mkdir -p config

    Pass/Fail tests

    • Create a script that returns exit status 0 for success.
    • Any non-zero value will be considered failure.
    • Script