https://github.com/milescranmer/gnn_resource_allocation

Code for our paper on doing resource allocation with graph neural networks
https://github.com/milescranmer/gnn_resource_allocation

astronomy astrophysics cosmology deep-learning graph-neural-networks inference machine-learning neural-network pytorch resource-allocation

Last synced: 18 days ago
JSON representation

Code for our paper on doing resource allocation with graph neural networks

• Host: GitHub
• URL: https://github.com/milescranmer/gnn_resource_allocation
• Owner: MilesCranmer
• License: apache-2.0
• Created: 2021-05-08T02:49:53.000Z (about 4 years ago)
• Default Branch: master
• Last Pushed: 2021-06-24T09:03:54.000Z (almost 4 years ago)
• Last Synced: 2025-05-01T12:48:11.061Z (18 days ago)
• Topics: astronomy, astrophysics, cosmology, deep-learning, graph-neural-networks, inference, machine-learning, neural-network, pytorch, resource-allocation
• Language: Python
• Homepage:
• Size: 8.12 MB
• Stars: 31
• Watchers: 3
• Forks: 2
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.txt

Awesome Lists containing this project

README

        
# Unsupervised Resource Allocation with Graph Neural Networks

![](https://github.com/MilesCranmer/gnn_resource_allocation/blob/master/schematic.svg)

Check out the paper [here](https://arxiv.org/abs/2106.09761).

PyTorch code for the forward model of our algorithm can be found in this repository in the file `model.py`. To train the model, execute `train.py`.

Data required to train this model can be found [here](https://app.globus.org/file-manager?origin_id=75a68b36-a6c0-11eb-92d8-6b08dd67ff48&origin_path=%2F)

Requirements for our codebase can be found in `environment.yml`. Note that one needs to use the following custom astropy:

```

pip install git+https://github.com/MilesCranmer/astropy

```

(it has some of the Cosmology calculations vectorized).

If you are using `conda`, and have CUDA version 11.0 and cuDNN version 8.0, you can create a duplicate of our env, using:

```bash

./create_env.sh gnn_allocation

```

which will create a new environment called `gnn_allocation`. This uses PyTorch 1.7.1, though it is likely to work for other versions if you decide to modify `create_env.sh` and `environment.yml`. You can also use an implementation without CUDA using the `environment_nocuda.yml` file.