How do I generate predictions on new data using a previously trained model?

You can either use Kumo’s REST API to automatically re-run pQuery training and batch predictions on a regularly scheduled basis, or retrain your pQuery and re-run your batch predictions directly in the Kumo SaaS UI. Both will allow you to make regular predictions on your unseen enterprise data. If you need to run inference on a different graph than the one used during training, see How do I generate predictions with a different graph than my training graph?.

Why is my PQuery underperforming for a particular subset of data?How can I diagnose problems with my data pipeline at batch prediction time?

⌘I

Documentation Index