Feature Request: Multilayer Hypergraphs with Heterogeneous Layer Types + Directed Layer Visualization

[ch8n57]

I would like to propose extending the framework to support multilayer hypergraphs with heterogeneous layer types, where each layer can independently be either directed or undirected. While the current implementation supports visualization for directed and undirected hypergraphs individually, the multilayer visualization is limited to only undirected layers. It would be valuable to enable multilayer visualization that can incorporate directed hypergraph layers alongside undirected ones, allowing mixed-layer representations within a single model. This enhancement would improve the expressiveness and applicability of the framework for real-world scenarios where different types of relationships (e.g., citation vs. co-authorship) coexist.

[leotrs]

Thanks for the suggestion! Multilayer hypergraphs with mixed directed/undirected layers is an interesting use case.

Would you be interested in putting together a PR for this? We'd be happy to review it and help you through the process. A good starting point would be looking at how draw_multilayer works in xgi/drawing/draw.py and how directed hypergraph drawing is handled separately.

If you have questions about the codebase or want to discuss the design before diving in, feel free to open a discussion or comment here.