Azure ML Loops through the different tasks

Question

I've got the following data structure in an Azure DB Table:

Client_ID | Customer_ID | Item | Preference_Score

The table can contain different datasets from different clients but the data structure is always the same. Then, the table is imported in Azure ML.

What I need is to repeat the same sequence of tasks in Azure ML for all the Client_ID in the above mentioned table.

So that in the end I will train a single model for each client and score the data of each single client individually and append the scored data and store it again in Azure SQL.

Is there any for each task in Azure ML like in SSIS? What's the best way to do this?

Thanks.

Answer 1

You can start from Azure Data Factory for automating batch scoring. In your model instead of web service output, you can use DataWriter exporter module to write the output directly into an Azure Table etc. You can check Microsoft MyDriving reference guide (http://aka.ms/mydrivingdocs) at page 107-8 where the machine learning section starts at page 100.