---

MuonTriggerPhase2RPC: python code for toy simulation of ATLAS RPC trigger

This package contains python macros for toy simulation of the ATLAS RPC trigger and for study of neural network regression for estimating transverse momentum of muon candidates.

Dependencies

These macros require Python 3.7+ and PyTorch

Documentation

Make events for NN training without noise:

python3 macros/runMuonSimulation.py --logy -n 100000 --noise-prob=0.0 --seed=42

Run NN training using these events:

python3 macros/trainNN.py rpc-sim_100000-events_0000-noise_000042-seed/events.csv

Make events for NN evaluation with noise:

python3 macros/runMuonSimulation.py --logy -n 200000 --noise-prob=0.001 --seed=101042

Make plots:

python3 macros/runMuonSimulation.py --logy --in-pickle=rpc-sim_200000-events_0010-noise_101042-seed/events.pickle --torch-model=rpc-sim_100000-events_0000-noise_000042-seed/events.pt --plot

Draw event displays:

python3 macros/runMuonSimulation.py --logy --in-pickle=rpc-sim_200000-events_0010-noise_101042-seed/events.pickle --torch-model=rpc-sim_100000-events_0000-noise_000042-seed/events.pt --draw