0

I have to do a sales prediction and I'm evaluation using a ML.NET solution hosted in a virtual machine(in Azure) vs using Azure ML Studio. The data may change once or twice per month. Which solutions should I choose? Also, for my use case, pricing might be a factor. Thank you.

Marius B.
  • 478
  • 6
  • 18

1 Answers1

1

In short: If you are building a .NET application and want to integrate ML, use ML.NET. If you don't do .NET, use Azure ML. Docs are helpful here: https://learn.microsoft.com/en-us/azure/architecture/data-guide/technology-choices/data-science-and-machine-learning

Manu Meyer
  • 191
  • 6
  • From a pricing perspective, for my use case, do you have any insights on which might be cheaper? – Marius B. Dec 06 '19 at 08:53
  • 1
    Well, Azure ML is a cloud service where you pay for the compute power that you "burn" whereas ML.NET is a Toolkit for .net that you can run anywhere. You don't pay anything for using ML.NET itself. – Manu Meyer Dec 08 '19 at 18:58