gcn4r's Introduction

GCN4R

Welcome to the GCN4R package, which is the first R package to allow custom fitting of Graph Neural Networks to social/biological network data. We are currently updating our documentation, but for now you can find a complete description of our work here: https://www.biorxiv.org/content/10.1101/2020.11.02.364935v1

To install:
First, install the python package:

pip install git+https://github.com/jlevy44/GCN4R

Install additional python dependencies:

install_gcn4r_dependencies ${TORCH_VERSION} ${CUDA_VERSION}

With TORCH_VERSION and CUDA_VERSION set according to https://pytorch-geometric.readthedocs.io/en/latest/notes/installation.html. Then, install the R package:

devtools::install_github("jlevy44/gcn4r")

Then, either read our tutorial paper or check out our R Markdown tutorial here to get started.

gcn4r's People

Contributors

Stargazers

Watchers

gcn4r's Issues

Fail to install on Windows

Hi,
Thanks for the great work, just what I am looking for.
However, I have problem when installing on windows OS.
here is the message

> devtools::install_github("jlevy44/GCN4R")
Error: Failed to install 'unknown package' from GitHub:
  HTTP error 404.
  Not Found

  Did you spell the repo owner (`jlevy44`) and repo name (`GCN4R`) correctly?
  - If spelling is correct, check that you have the required permissions to access the repo.

I am with Microsoft R open 4.0.2

Please advise,
Many thanks,
eric

By the way, what version of R would you recommend?

Add/Remove Self-Loops

Grad to Adjacency

Identifying nodes

Hi Joshua Levy,

First off thank you for such a wonderful contribution.
And in your demo file, also using the package i was able to develop plots with nodes and edges.
Is there any way to know which nodes are highly linked using gcn4r, or could we find out the node id or name of the node from the plot, so that we can better infer that this node is related to that node. I tried to make the node interactive but it didnt work out well.
Do you happen to have some solution??
Thanks in advance.