News for hackers. Stuff that matters.

StellarGraph

StellarGraph v0.11 open-source Machine Learning Library for graphs released

Open-source Python library with state-of-the-art graph ML algorithms delivered by CSIRO’s Data61.5 new algorithms + streamlined onboarding are the features of StellarGraph 0.11, with substantial API, documentation + demo improvements. Full release notes: https://github.com/stellargraph/stellargraph/releases/tag/v0.11.0New algorithms * Watch Your Step: computes node embeddings by simulating the...

( Read More... | no comments | 3 points | on HN )
Hi @mooneater. At the moment, the library is focused on graphs that fit in memory. Managing out-of-core data brings a whole extra layer of caching, data transfer and data management complexity; we prefer to focus on extending and increasing the algorithms we support. We think this is the appropriate trade-off because many problems involve relatively small graphs, and, even for larger ones, it’s getting easier and cheaper to rent machines with hundreds of gigabytes or even terabytes of RAM. That said, we do have experimental support for running GraphSAGE against a graph stored in Neo4j, and we’re working towards better support for Neo4j in general, e.g. https://github.com/stellargraph/stellargraph/issues/1034.

StellarGraph v0.10 Open-Source Python Machine Learning Library for Graphs

StellarGraph is an open-source library featuring state-of-the-art graph machine learning algorithms. The project is delivered as part of CSIRO’s Data61.Dramatically improved memory usage is the key feature of the 0.10 release of the library, with the StellarGraph and StellarDiGraph classes now backed by NumPy and Pandas. This will enable significant performance benefits.Version 0.10...

( Read More... | 3 comments | 19 points | on HN )

StellarGraph v0.8.1 Open-Source Python Machine Learning Library for Graphs

StellarGraph is an open-source library implementing a variety of state-of-the-art graph machine learning algorithms. The project is delivered as part of CSIRO’s Data61.We are happy to announce the 0.8.1 release of the library, which extends StellarGraph capability by adding new algorithms and demos, enhancing interpretability via saliency maps for Graph Attention (GAT), and further simplifying...

( Read More... | no comments | 5 points | on HN )