Use the Azure Machine Learning activity to run a job on an Azure Machine Learning instance

The Azure Machine Learning activity in Data Factory for Microsoft Fabric allows you to run a job on an Azure Machine Learning instance.

Prerequisites

To get started, you must complete the following prerequisites:

• You must have access to a Microsoft Fabric tenant with a provisioned capacity. You can try Fabric with a free trial.
• A Fabric workspace assigned to that capacity.

Add an Azure Machine Learning activity to a pipeline with UI

To use an Azure Machine Learning activity in a pipeline, complete the following steps:

Create the activity

1. Create a new pipeline in your workspace.

2. Search for Azure Machine Learning in the pipeline Activities pane, and select it to add it to the pipeline canvas.

   Note

   You may need to expand the menu and scroll down to see the Azure Machine Learning activity as highlighted in following the screenshot.

   

3. Select the new Azure Batch activity on the pipeline editor canvas if it isn't already selected.

   

Refer to the General settings guidance to configure the General settings tab.

Azure Machine Learning activity settings

1. Select the Settings tab, then you can choose an existing or create a new Azure Machine Learning connection.
2. Choose and Endpoint type, either Batch Endpoint or Pipeline (v1).
3. Provide a Batch endpoint and Batch deployment and configure **Job settings for the Batch Endpoint type, or provide the pipeline details to run an Azure Machine Learning Pipeline (v1).

Save and run or schedule the pipeline

Switch to the Home tab at the top of the pipeline editor and select the save button to save your pipeline. Select Run to run it directly or Schedule to schedule runs at specific times or intervals. For more information on pipeline runs, see: schedule pipeline runs.

After running, you can monitor the pipeline execution and view run history from the Output tab below the canvas.

Known issues

• Using Service Principal to run a notebook that contains Semantic Link code has functional limitations and supports only a subset of semantic link features. See the supported semantic link functions for details. To use other capabilities, you're recommended to manually authenticate semantic link with a service principal.
• Azure Machine Learning (AML) activity may fail in some configurations due to a missing dual‑token audience during authentication. The fix is currently being worked on.

Related content

• How to monitor pipeline runs