Skip to main content

Batch Inference

You'll score data with a registered model — using pandas for quick checks or Spark UDFs for production scale — in ~10 min.

Prereqs: Save a Model to Unity Catalog

What you'll build

A batch job that loads a model from Unity Catalog by alias, scores rows, and writes predictions to a Delta table.

Prerequisites

  • A model registered in Unity Catalog (see Save a Model to Unity Catalog).
  • Data available in a Delta table or DataFrame to run inference against.

Inference methods

MethodBest forReference
🥇 Spark UDFProduction-scale tabular data distributed across a clusterHow to speed up inference
Pandas single-nodeSmall batches, quick validation, single-node notebooks
Pandas UDF for imagesComputer vision workloads with images stored in object storageDistributed inference with pandas UDF

Next