AnalysisConfigs

Repository containing analysis configurations for PocketCoffea

Setup

The first step is installing the main PocketCoffea package in your python environment.

Please have a look at the Installation guide.

The configs package has been created to separate the core of the framework from all the necessary configuration files and customization code needed the different analyses. The configuration is structured as a python package to make easier the import of customization code into the framework configuration and also to make the sharing of analysis code easier.

Once you have a PocketCoffea local installation, you can install the configs and utils package with:

cd AnalysisConfigs
pip install -e .

This will install the configs package in editable mode.

Analysis examples

• Simple Z-> mumu invariant mass analysis here

Run analysis

run_pocket_coffea <config_name> <config_file> <run_options> <output_dir> <--test>

Run ggF HH4b analysis

cd configs/HH4b
pocket-coffea run --cfg HH4b_parton_matching_config.py -e dask@T3_CH_PSI --custom-run-options params/t3_run_options_spanet_predict.yaml -o /work/mmalucch/out_test --executor-custom-setup onnx_executor.py

Run VBF HH4b analysis

cd configs/VBF_HH4b
pocket-coffea run --cfg VBF_HH4b_test_config.py -e dask@T3_CH_PSI --custom-run-options params/t3_run_options_spanet_predict.yaml -o /work/mmalucch/out_hh4b/out_vbf_jets_candidates/  --executor-custom-setup onnx_executor.py

Run plot 2bvs4b

sbatch -p short --account=t3 --time=00:05:00 --mem 25gb --cpus-per-task=8 --wrap="python plot_2bMorphedvs4b.py -i <input_directory> -o <output_directory>"

Run plot DNN_score

sbatch -p short --account=t3 --time=00:10:00 --mem 40gb --cpus-per-task=1 --wrap="python ~/AnalysisConfigs/scripts/plot_DNN_score.py -id ./  -im output_GluGlutoHHto4B_spanet_kl-1p00_kt-1p00_c2-0p00_2022_postEE.coffea -r2 -om /work/mmalucch/out_ML_pytorch/DNN_DHH_method_class_weights_e5drop75_postEE_allklambda_matteo/state_dict/model_best_epoch_19.onnx"