• Identify the differences and advantages of model serving approaches: batch, realtime, and streaming

• Deploy a custom model to a model endpoint

  Deploy a Custom Model

• Deploy a model to a model endpoint

  • Log the model or code in the MLflow format, either native MLflow built-in flavors or pyfunc
  • After the model is logged, register it in the Unity Catalog or workspace registry
  • From here, you can create a model serving endpoint to deploy and query your model

• Use pandas to perform batch inference

  • Batch Deployment

• Identify how streaming inference is performed with Delta Live Tables

  • Streaming/Pipeline Deployment

• Deploy and query a model for realtime inference

• Split data between endpoints for realtime interference

Summary

Covers model deployment methods (batch, real-time, streaming), deploying custom models to endpoints, using MLflow for logging and registering models, batch inference with pandas, streaming inference with Delta Live Tables, and real-time model querying and data splitting.