Water Cherenkov Machine Learning for neutron identification with graph neural nets

Description

Python implementation of the training engine and framework to build, train and test graph neutral net models for Water Cherenkov Detectors.

Table of Contents

1. Directory Layout

2. Installation

Directory Layout

.
+-- visualisation                       # Tools for visualizing model performance and dataset features
  +-- detector_geometry.ipynb
  +-- full_geo_dump.npz
  +-- mPMT_to_row_col.csv
  +-- visualisation.ipynb
+-- root_utils                          # Tools for interacting with the ROOT files from the WCSim simulations
  +-- pos_utils.py
+-- README.md                           # README documentation for the repository

Installation

Requirements

The following Python standard, machine learning and deep learning libraries are required for the functionality of the framework :

1. PyTorch
2. PyTorch Geometric
3. NumPy
4. Matplotlib

To download the repository use :

git clone https://github.com/WatChMaL/NeutronGNN.git