Skip to content

saalfeldlab/decomp-gnn

Repository files navigation

Decomposing heterogeneous dynamical systems with graph neural networks

Cédric Allier, Magdalena C. Schneider, Michael Innerberger, Larissa Heinrich, John A. Bogovic, Stephan Saalfeld

Janelia Research Campus, Howard Hughes Medical Institute

Summary of the GNN workflow

Setup

Run the following line from the terminal to create a new environment 'gnn':

conda env create -f environment.yaml

Activate the environment:

conda activate gnn

Then, you should be able to import all the modules from the package in python:

from ParticleGraph import *

To create the paper's figures run the scripts in decomp_gnn/paper_experiments.

Citation

@article{allier2024decompgnn,
  author = {Allier, Cédric
            and Schneider, Magdalena C.
            and Innerberger, Michael
            and Heinrich, Larissa
            and Bogovic, John A.
            and Saalfeld Stephan},
  title = {Decomposing heterogeneous dynamical systems with graph neural networks},
  archiveprefix = {arXiv},
  eprint = {2102.04360},
  year = {2024}
}

About

Decomposing heterogeneous dynamical systems with graph neural networks

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •  

Languages