Amazon Neptune ML is now generally available with support for edge predictions, automation, and more

Amazon Neptune ML, a machine learning capability for graphs that uses Graph Neural Networks (GNNs), is now generally available in all regions where Amazon Neptune is available. Using the Deep Graph Library (DGL), an open-source library to which AWS contributes, Neptune ML automates the heavy lifting of selecting and training the best ML model for graph data. With Neptune ML, you can improve the accuracy of most predictions for graphs by over 50% when compared to making predictions using non-graph methods.

Graph applications such as product recommendations, knowledge graphs, customer 360, and fraud detection can provide deep insights with information such as predicted labels on new vertices and edges. You can now use Neptune ML to predict properties or numerical values on edges using classification and regression models. Additionally, Neptune ML automates feature encoding and instance selection based on the graph data and size, making it even easier for customers to run machine learning workloads on Neptune. For changes to your graph data, you can also update predictions by using existing Neptune ML models or by retraining using the previously identified best hyperparameters in existing models.

To get started, use a Neptune ML quick-start CloudFormation stack to walk through pre-built tutorials in Neptune notebooks. To learn more about the new enhancements, see the Neptune ML documentation page. Neptune ML is available to customers for production use starting in Neptune version 1.0.5.0. You only pay for the resources provisioned such as Amazon Neptune, Amazon SageMaker, Amazon CloudWatch, and Amazon S3.