# HexagDLy - Processing hexagonally sampled data with CNNs in PyTorch

## TL;DR

HexagDLy is a Python library that extends PyTorch to enable convolution and pooling operations on hexagonally sampled data, facilitating deep learning applications in fields like astroparticle physics.

## Contribution

It introduces new convolutional and pooling layers for hexagonal grids, making CNNs more accessible for hexagonally sampled data.

## Key findings

- Enables effective processing of hexagonally sampled data with CNNs
- Simplifies implementation of deep learning models for specialized data formats
- Supports applications in astroparticle physics and related fields

## Abstract

HexagDLy is a Python-library extending the PyTorch deep learning framework with convolution and pooling operations on hexagonal grids. It aims to ease the access to convolutional neural networks for applications that rely on hexagonally sampled data as, for example, commonly found in ground-based astroparticle physics experiments.