cetin
7c2194df23
and GEC Filter to eff options files trained network with correct parameterisation sample4 new effs with sample4 NN weights |
||
---|---|---|
arc6_data | ||
data_matching | ||
env | ||
moore_options | ||
neural_net_training | ||
nn_neural_net_training | ||
parameterisations | ||
scripts | ||
thesis | ||
.gitignore | ||
.gitlab-ci.yml | ||
.pre-commit-config.yaml | ||
electron_main.py | ||
LICENSE | ||
main_tracking_losses.py | ||
main.py | ||
README.md | ||
setup.sh | ||
tuner.code-workspace |
Parameterisation Tuner
This project provides utils for producing magic parameters used by the pattern recognition algorithms in the Rec project. Typical parameters are coefficients for extrapolation polynomials and weights for TMVA methods. This is based on this repo by André Günther.
Setup
There's a bash script for setting up the necessary (python) environment. Simply do:
chmod +x setup.sh
./setup.sh
This will install dependencies like ROOT and Jupyter. To enter the environment do:
source env/tuner_env/bin/activate
conda activate tuner