TheBatchPredictionJob resource lets you run an asynchronousprediction request. Request batch predictions directly from themodelresource. You don't need to deploy the model to anendpoint. For data typesthat support both batch and online predictions you can use batch predictions.This is useful when you don't require an immediate response and want to processaccumulated data by using a single request.

To make a batch prediction, specify an input source and an output locationfor Vertex AI to store predictions results. The inputs and outputsdepend on themodel type that you're working with. For example, batchpredictions for the AutoML image model type require an inputJSON Lines file and the name of a Cloud Storage bucket to store the output.For more information about batch prediction, seeGet batch predictions.

You can use theModelBatchPredictOp component to access this resource through Vertex AI Pipelines.

API reference

• For component reference, see theGoogle Cloud SDK reference for Batch prediction components.
• For Vertex AI API reference, see theBatchPredictionJob resource page.

Tutorials

• Custom training with prebuilt Google Cloud Pipeline Components

Version history and release notes

To learn more about the version history and changes to the Google Cloud Pipeline Components SDK, see theGoogle Cloud Pipeline Components SDK Release Notes.

Technical support contacts

If you have any questions, reach out tokubeflow-pipelines-components@google.com.