GDS 1.4.0 (Graph Native Machine Learning!) is live!

Hello there! We've just GA'd GDS 1.4 (after the preview release was launched on the 20th)

It's been posted to Github and the Download Center, and should be available in Desktop as well. Full release notes are available here, but this includes all the awesome work on embeddings , KNN, license keys, and more -- including bug fixes for lots of existing algorithms.

If you haven't seen it yet, our blog post provides a great summary of what's new and exciting :tada:

Is there a pointer to what this is all about? (A quick search couldn't find anything that was that helpful...)

I know ML pretty well and I'm coming up to speed on Graph DB's. I just don't quite get what the two of them together can do in a concrete example. (I saw the drug discovery example, but it wasn't clear to me what was being found or how.)

(added) I did find this: https://neo4j.com/docs/graph-data-science/current/

I think this could have a lot of potential, but I can't quite grasp it yet.

Thanks in advance!

Hi @clem - check out our intro to graph algorithms book: https://neo4j.com/graph-algorithms-book/

That provides a great overview of what graph algorithms are and why they're useful. You may also want to check out our developer guides, which provide overviews of specific concepts, how to guides, and tutorials: https://neo4j.com/developer/graph-data-science/