We wrote a little something! @nwlandry.bsky.social and I talk about how new software to study networks with groups (like @xgi.bsky.social !) can help interdisciplinary collaborations tackle old questions from social ontology and the science of groups.
www.siam.org/publications...
@xgi.bsky.social in the wild 👀
@xgi.bsky.social in the wild 👀
@nwlandry.bsky.social fighting against vibes-based estimations of the convergence when sampling from the configuration model. Good content.
Bonus: @xgi.bsky.social viz
Tim LaRock talking about the number (and properties) of hypergraphs with the same pairwise projection. Very cool work and a picture of @xgi.bsky.social in the wild 👀
Hypergraph Link Prediction via Hyperedge Copying
https://arxiv.org/abs/2502.02386
We propose a generative model of temporally-evolving hypergraphs in which hyperedges form via noisy copying of previous hyperedges. Our proposed model reproduces several stylized facts from many empirical hyperg...📈🤖
XGI in the wild!
New release! Version 0.9.5 adds support for hyperedge line width. We also fixed an error in our random generative models and made some methods up to 4x faster (!!!) in the process. @kaiserdaniel.bsky.social also made his first contribution, fixing an error in the hypergraph Laplacian. Thanks Daniel!
XGI in the wild!
Did you know that XGI can load chemical reaction datasets directly from the BiGG repository? bigg.ucsd.edu. Simply run "xgi.load_bigg_data()" and see what datasets are available!
We have coauthorship networks, email networks, contact networks, ecological networks, biological networks and more!
Did you know that XGI offers 39(!) higher-order datasets in a standardized format? xgi.readthedocs.io/en/stable/xg...
Guess what? Now, we have benchmarks for core functionality! Any suggestions for additional benchmarks? xgi-org.github.io/xgi/dev/bench/
🙋
Want a feature that you don’t see listed here? Just give us a shout, and we will do our best to make it a reality! As always, contributions are always welcome!
Finally, we have almost DOUBLED the number of higher-order datasets that XGI offers (24). In addition, we have a summary table of the network statistics (xgi.readthedocs.io/en/stable/xg...) so you can quickly find a dataset that meets your needs.
Thanks to @nwlandry.bsky.social for this!
🎨 Visualization
Drawing has gotten more flexible thanks to Maxime Lucas: you can now specify almost any drawing parameter that matplotlib does!
Stay tuned for a longer thread about this soon, and go check the tutos: xgi.readthedocs.io/en/stable/ap...
📖 Tutorials
- We've organized them by type (getting started, focus, and in-depth)
- The tutos are now integrated to the online docs: xgi.readthedocs.io/en/stable/ap...
- We've also got recipes, go check them out: xgi.readthedocs.io/en/stable/ap...
Big thanks to Thomas Robiglio for this!
X-mas has come early this year for XGI-ers! We are excited to announce massive changes to our documentation, datasets, and functionality!
This follows our latest poll (below): see thread below for details 👇🧵
This is the bluesky account for XGI, a Python library for working with higher-order networks! Expect periodic updates on what we're working on, new releases, and research that uses XGI. We're excited to engage with the community here!
Hello, hypergraphy world!
