Skip to content

prayagyadav/coffea-fcc-analyses

Repository files navigation

COFFEA FCC Analyses

Setup

Start a singularity shell with coffea with

./shell

Examples

Example Analyses are separated into these files:

  • processor_<example_name>.py
  • runner_<example_name>.py
  • plotter_<example_name>.py

To run the ZH Recoil example:

Step 0 : Clone this repository

  1. Navigate to example folder

    cd examples/FCCee/higgs/mHrecoil/mumu/
  2. Start the singularity shell

    ./shell
  3. A] Local Execution Within the shell execute runner_mHrecoil.py with the desired parameters(--executor or -e set to dask by default). The output file would be saved at ./output/FCCee/higgs/mH-recoil/mumu/.

    singularity> python3 runner_mHrecoil.py -e dask

    One can also choose to split up the processing and results in chunks with the --chunks or -c argument

    singularity> python3 runner_mHrecoil.py -e dask -c 8

    B] Batch Execution with HTCondor Choosing the executor as condor one can generate job and submit files and a master submit file called condor.sh

    singularity> python3 runner_mHrecoil.py -e condor -c 8

    Now exit the container shell with exit , move to the directory where the files are created(by default Batch) and run condor.sh without getting into the shell

    singularity> exit
    cd Batch
    ./condor.sh
  4. Generate plots with plotter_mHrecoil.py (from the singularity shell). The plots would be saved at ./output/FCCee/higgs/mH-recoil/mumu/plots/

    singularity> python3 plotter_mHrecoil.py

    If the plots are generated in a different folder than ./output/FCCee/higgs/mH-recoil/mumu/ , input it's path with the --input or -i keyword. For example for batch submission, the outputs are saved in Batch directory.

    singularity> python3 plotter_mHrecoil.py -i Batch