How to calculate features in future time frames in azure ml

Alex B 1 Reputation point
2019-11-06T12:02:01.267+00:00

Hello,

I'm new to azure and have run into a problem when deploying my first experiment. Suppose I'm trying to forecast the retail sales for a location, the features used are the historical weather and foot traffic. It seems that if I want to forecast the sales for a week into the future, I will have to provide the weather and foot traffic expected at that time, which means that I will need to forecast the features as well. Is this correct? What am I missing?

Thank you

Azure Service Fabric
Azure Service Fabric
An Azure service that is used to develop microservices and orchestrate containers on Windows and Linux.
269 questions
0 comments

1 answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. ajkuma 26,471 Reputation points Microsoft Employee
    2019-11-07T12:26:32.343+00:00

    @Alex B ,

    Thanks for posting the question on Microsoft Q&A.

    Azure Machine Learning service is yet to be migrated to Q&A. Please post your question directly on the MSDN forum for the service (Azure Machine Learning forum)

    We are actively working to onboard all services. We will make a public announcement once complete.
    Thank you for your patience as we evolve the Q&A platform.

    Kindly re-post the question on the ML forum on MSDN with more details on your scenario and any error you may have seen with your first experiment.
    Having mentioned that, I see a good document 'Auto-train a time-series forecast model' which describes on how to prepare the feature data for forecasting. Hope that helps!

    You may wish to see the article: Microsoft Q&A - Frequently Asked Questions

    1 person found this answer helpful.
    0 comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.