Skip to content

Click in-app to access the full platform documentation for your version of DataRobot.

Batch prediction jobs

You might want to make a one-time batch prediction but you might also want to schedule regular batch prediction jobs. This section shows how to create and schedule batch prediction jobs.

Be sure to review the predictions interface considerations before proceeding.

Create a prediction job definition

Job definitions are flexible templates for creating batch prediction jobs. You can store definitions inside DataRobot and run new jobs with a single click, API call, or automatically via a schedule. Scheduled jobs do not require you to provide connection, authentication, and prediction options for each request.

To create a job definition for a deployment, navigate to the Predictions > Job Definitions tab. The tab lets you configure the details required by the batch prediction API. The following table describes the information and actions available on the New Prediction Job Definition tab.

Field name Description
Prediction job definition name Enter the name of the prediction job that you are creating for the deployment.
Prediction source Select the intake source type and define the connection for where the data to be scored comes from.
Prediction options Configure the prediction options as outlined in step 2 of the make predictions procedure.
Prediction destination Indicate the output destination for predictions. Select the destination type and define the connection.
Jobs schedule Toggle whether to run the job immediately and whether to schedule the job.
Save and run prediction job definition Click this button to save the job definition and run the job now, schedule it to run later, or both.

Once fully configured, click Save prediction job definition.

Schedule prediction jobs

You can schedule prediction jobs to run automatically on a schedule. When outlining a job definition, toggle the jobs schedule on. Specify the frequency (daily, hourly, monthly, etc.) and time of day to define the schedule on which the job runs.

For further granularity, select Use advanced scheduler. You can specify the exact time for the prediction job to run, down to the minute.

Set up intake and output adapters

You can switch between different options for intake and output adapters by picking an option through the Source Type and Destination Type dropdown menus. You must complete the appropriate authentication workflow for each type.

Database connections

  • JDBC

Cloud Storage Connections

  • Azure
  • GCP
  • S3

Data Warehouse Connections

  • BigQuery
  • Snowflake
  • Synapse

Other Connections

  • AI Catalog

Database connections

  • JDBC

Cloud Storage Connections

  • Azure
  • GCP
  • S3

Data Warehouse Connections

  • BigQuery
  • Snowflake
  • Synapse

Other Connections

  • Tableau

For information about supported data sources, see Data sources supported for batch predictions. See also the intake and output options documentation.

Run a prediction job

After creating a prediction job, you can run it by clicking from the menu to the right of the job definition and selecting Run now. After running it, you will receive a notification that takes you to the Prediction Jobs page where you can follow the progress of the job.

View prediction jobs

To view prediction jobs, navigate to Deployments > Prediction Jobs. You can view all the jobs that are currently running or have already completed. Any other prediction made on a deployment appears on this page. You can filter the jobs by status, start and end time, deployment ID, and type.

For jobs that are currently running, you can view the logs in progress (if the job does not use streaming intake), cancel the job, or go to the deployment or job definition. For jobs that have finished running, you can also run the job again.

The following table describes the information and actions available from the Prediction Jobs list.

Category Description
Job Definition The job definition used to create the prediction job.
Added to Queue Time at which the prediction job was initialized.
Created by User who triggered the job.
Status State of the job.
Source Intake adapter for this prediction job.
Destination Output adapter for the prediction job.

For more details on the fields above, reference the Batch Prediction API documentation.


Updated December 4, 2021
Back to top