How to calculate features on forecasted time frame RRS feed

  • Question

  • 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

    Wednesday, November 6, 2019 3:23 AM

All replies

  • Hi,

    Can you please share the code/doc link that you are trying and error details to debug it further.

    Yes, we need to forecast the features. Please follow the following forecasting samples available in AutoML.

    Please follow the below documents for feature engineering with the time series data.




    Friday, November 8, 2019 2:30 AM
  • Hi,

    You’d be providing the features to train the model and forecasting sales value. Please feel free to check out various forecasting examples on Azure AI Gallery. This example is a good starting point. Hope this helps. Please let me know if you have any more questions. Thanks.



    If a post helps to resolve your issue, please click “Mark as Answer” and/or “Vote as helpful”. By marking a post as Answered and/or Helpful, you help others find the answer faster.  Thanks.

    Friday, November 8, 2019 2:31 AM
  •  Hello,

    The example link provided contains detail explanation which is fine. But it has not provided guidelines to predict the feature values. To get in more details, consider the below dataset and consider today is: 2019-11-11. I have last 2 years of daily data and below is last 6 rows:

    Date, Temperature, Sales

    2019-11-06, 25.5, 500000

    2019-11-07, 24.2, 550000

    2019-11-08, 25.1, 560000

    2019-11-09, 22.6, 510000

    2019-11-10, 22.3, 520000

    2019-11-11, 24.4, 535000

    ------- Now I have to predict Sales for 2019-11-12, 2019-11-13, 2019-11-14. In order to predict sales for those dates, I have to provide below test data to the machine learning trained model:

    Date, Temperature

    2019-11-12, temperatureX

    2019-11-13, temperatureY

    2019-11-14, temperatureZ

    -- What will be values for temperatureX, temperatureY and temperatureZ since these values will be coming from future as well ?



    Monday, November 11, 2019 10:07 AM