Particle Identification Challenge

Assignment 2, COMP9417: Kaggle competition (https://www.kaggle.com/c/trackml-particle-identification). The challenge is to build an algorithm that quickly reconstructs particle tracks from 3D points left in the silicon detectors.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for testing purposes.

Prerequisites

• Python 3.0+ either locally or in a virtual environment.
• Datasets from kaggle: https://www.kaggle.com/c/trackml-particle-identification/data
• Packages required:
  • trackml: https://github.com/LAL/trackml-library.git This is the competition-specific package.
  • numpy, pandas, tqdm, scikit_learn for our current best-performing solution train_trackml-DBSCAN.py (see the next section for how to install).
  • The other version, train_trackml-HDBSCAN.py, which gives a lower score than the DBSCAN version, requires an extra package hdbscan. However this package is not able to be installed on CSE lab computers unless with sudo permission.

Installing

First, clone this repository to your local machine. You may skip this step if you already have all the files: two python scripts and a requirements.txt.

git clone https://github.com/XifeiNi/TrackML.git

There is a requirements.txt bundled in this repository that lists the required packages you need to install. These are all compatible with the CSE lab environment. They can be installed using the following commands:

pip3 install --user -r requirements.txt

If that fails, you may use the following commands:

pip3 install --user git+https://github.com/LAL/trackml-library.git
pip3 install --user numpy pandas scikit_learn tqdm

Additionally, if you wish to run the alternative solution train_trackml-HDBSCAN.py, you will need to install one more package using the following command:

pip3 install --user hdbscan

However note that this package is not able to be installed on CSE lab computer unless with sudo permission, because it requires certain linux packages that are missing on the lab computers. As it is not required in our current best-performing solution, we have not included it in requirements.txt.

Running

Run either script as usual:

python3 train_trackml-DBSCAN.py

python3 train_trackml-HDBSCAN.py

Authors

• Cecilia Ni
• Shahedul Islam
• Kavi Shah
• Yi Xiao

License

This project is licensed under the MIT License - see the LICENSE.md file for details

Acknowledgments

• The kaggle community
• CERN: the European Organization for Nuclear Research