Azure Machine Learning Linked Service

  • Article

 

Important

This topic provides the JSON format that is supported by the older versions of Azure PowerShell. If you are using the July 2015 Release version of Azure PowerShell or later, see Azure Machine Learning Linked Service for the latest JSON format. You can convert the JSON from old format to new format by using the tool: JSON Upgrade Tool

You create an Azure Machine Learning linked service to link an Azure Machine Learning model to an Azure data factory. This topic describes the JSON properties for defining an Azure Machine Learning linked service. When you use the Data Factory Editor in Azure Portal, the editor provides you a JSON template for defining a linked service that you can modify and deploy to create a linked service. To use the T:Microsoft.Azure.Commands.DataFactories.New-AzureDataFactoryLinkedService cmdlet to create a linked service, you will need to create a JSON file and use it with the cmdlet.

Important

This topic provides the JSON format that you can use to define an Azure Machine Learning linked service with old versions of API (prior to 2015-05-01-preview). If you are using the newer/latest version of API, see Azure Machine Learning Linked Service for the new/latest JSON format.

Properties

Property

Description

type

The type property should be set to: AzureMLLinkedService.

mlEndpoint

The batch scoring URL.

apiKey

The published workspace model’s API.

Example

{
    "name": "MyAzureMLLinkedService",
    "properties":
    {
        "type": "AzureMLLinkedService", 
        "mlEndpoint":"https://[batch scoring endpoint]/jobs", 
        "apiKey":"[apikey]" 
    } 
}

Walkthrough

See Create Predictive Pipelines using Azure Data Factory and Azure Machine Learning article for details and a walkthrough.