/ElectromagneticShowersReconstruction

Reconstruction and generation of 3D shower structures for neutrino experiments

Primary LanguageJupyter NotebookMIT LicenseMIT

Reconstruction of 3D Shower Structures for Neutrino Experiments

Authors: [V. Belavin] (vbelavin@hse.ru), [E. Trofimova] (etrofimova@hse.ru), [A. Ustyuzhanin] (austyuzhanin@hse.ru)

Overview

This directory contains code necessary to run the Electromagnetic Showers (EM) Reconstruction algorithm that is devided into the following parts:

  1. Graph Construction;
  2. Edge Classification;
  3. Showers Clusterization;
  4. Parameters Reconstruction.

Experimental Data

X, Y, Z coordinates and the direction of the EM Showers base-tracks.

The showers are generated using FairShip framework.

Data for graph generation is located here: https://gitlab.com/SchattenGenie/shower_generation/blob/master/data/mcdata_taue2.root

Results

The algorithm detects ~ 86% of Showers and assess the coordinates and direction of base-tracks with ~ 75% accuracy.

Clusters Examples: Clusters Examples

Running the code

training_classifier.py predicts the probability of edge to connect vertices of one shower. The probability will be then used as edge weight for proposed clusterization algorithm in clustering.py.