Azure ML Batch Execution Task
The "Azure ML Batch Execution Task" is used to execute a batch job for an Azure ML web service. See the Azure Storage Connection Manager page to learn more about setting up the connection manager.
Azure ML Batch Execution is available for SQL versions 2012 and higher.
• Storage Container - Identifies the container within Azure storage where the input and output files are stored.
• Input File - The name of the source file located in the Azure storage container.
• Output File - The name of the destination file located in the Azure storage container (does not need to exist.)
• Web Service Output Name - The name of the web service output defined in the Azure ML experiment web service.
• Base Web Service Url - The request Uri defined in the batch execution api documentation for your web service. To obtain this Uri, users will need to do the following in Azure ML Studio:
- Select the web service group
- Select the Batch Execution Url in the default endpoint
- Select the Request Uri until the end of the word "jobs" and copy (see screenshot below):
• Execution Timeout - The amount of time (in seconds) before the component will send a cancel command.
• Parameters - If users have a web service with parameters, this is where to define them.