With the migration to the new platform, it looks like we've lost the specific topic pages, and instead have to navigate by Product Area and labels. However, there are a bunch of semi-duplicate labels (eg. Graph Algorithms, Graph AlgorithmsGraph Data ...
Get excited! Today, we've launched:
GDS 2.0 - a massive update to the library we all know and love
AuraDS: graph data science... as a service
graphdatascience - our native python client
Want to learn more? Check out our release blog: Introducing Gra...
Our 1.7 release of GDS is now GA, and available on the download center, github, and desktop
The most exciting features in this release include:
Introduced Link Prediction Pipelines to make it easier to define and calculate features, split your graph...
The GDS 1.6 preview release has been published! Currently available on Github and our download center.There are a ton of new features in this pre-release, and you should read the release notes to see all of them, but some of the highlights are:
Degr...
Graph Data Science Library v1.5 now available on github, our download center, and desktop
This release includes:
ML based link prediction and node classification (train a machine learning model in Neo4j; includes appropriate graph splitting and m...
No, I don't think we don't expose the specific nodes in the hold out set. If this is a feature you need, you can create an issue on our github: https://github.com/neo4j/graph-data-science/issues
when you've trained a model, modelInfo and modelSelectionStats contains information about the trained model: https://neo4j.com/docs/graph-data-science/current/machine-learning/node-property-prediction/nodeclassification-pipelines/training/#nodeclassi...
I think this is probably an issue with browser formatting/displaying the streaming results for 15M nodes.If you want to check what the actual status of execution is, I would try gds.beta.listProgress to confirm that the operations finished. If you wa...
You can try the graph construct method in the python client: https://neo4j.com/docs/graph-data-science/current/python-client/graph-object/#_constructing_a_graphYou'd need to map your json to the correct pandas dataframe formatting, but that will let ...