abojchevski/graph2gauss

Gaussian node embeddings. Implementation of "Deep Gaussian Embedding of Graphs: Unsupervised Inductive Learning via Ranking".

47
/ 100
Emerging

This tool helps researchers and machine learning practitioners analyze complex network structures by transforming individual nodes within a graph into "Gaussian embeddings." You input a graph's adjacency matrix and optionally node features, and it outputs a statistical representation (mean and covariance) for each node that captures its position and uncertainty in the graph. This is useful for tasks like anomaly detection, link prediction, or visualizing relationships within networks.

179 stars. No commits in the last 6 months.

Use this if you need to represent nodes in a graph with probabilistic, low-dimensional vectors that capture both their location and uncertainty within the network structure.

Not ideal if you are looking for a simple, deterministic point embedding for each node without considering uncertainty or if you don't work with graph-structured data.

network-analysis graph-representation unsupervised-learning node-embedding data-science
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 21 / 25

How are scores calculated?

Stars

179

Forks

41

Language

Python

License

MIT

Last pushed

May 15, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/abojchevski/graph2gauss"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.