GraphNeT
GraphNeT: deep-learning (graph neural network) library for event reconstruction in neutrino telescopes
Python library for deep-learning-based reconstruction and classification of events in neutrino telescopes (IceCube, KM3NeT, P-ONE, …): data converters, graph representations of detector hits, GNN models and training pipelines, with pretrained models.
Get it
| Location | Role | Host | Formats | Size |
|---|---|---|---|---|
| https://github.com/graphnet-team/graphnet | primary | GitHub | Source code |
Metadata
| Creators | GraphNeT team |
|---|---|
| Owner | graphnet-team |
| Published | 2021-04-12 |
| Last updated | 2026-10-18 |
| Licence | Apache-2.0 |
| Access | Open |
| Experiments | IceCube, KM3NeT |
| Languages | Python |
| Purpose | Reconstruction |
| Dependencies | PyTorch, PyTorch Geometric |
| Development | active |
Used by
- Data sets NuBench: open benchmark datasets for deep-learning event reconstruction in neutrino telescopes
Know of another paper or code that uses this, or an input it depends on? Tell us.
Papers
- GraphNeT, J. Open Source Softw. 8, 4971 (2023) · doi:10.21105/joss.04971 · arXiv:2210.12194 (describes)
- GraphNeT 2.0 · arXiv:2501.03817 (describes)
How to cite
Curation
| Status | Curated (candidate: harvested, not reviewed · curated: reviewed metadata · verified: resource downloaded and checked against its paper) |
|---|---|
| Added | 2026-10-23 |