I am trying to build a model using the designer in Azure Machine Learning Studio that will need to predict multiple values simultaneously.
The scenario I am working with is there are a set of codes assigned to an order. When the order is audited, the codes may be adjusted (either added, removed, changed or a combination of all). What this results in is a before and after set of codes assigned to an order.
What I want to do is feed the before and after codes plus other characteristics of the order into a model so going forward it is possible to provide suggestions to the auditor on potential changes to 'before' codes to create the 'after' codes.
I am wanting to understand if this can be done using the designer or if it would need to be done using code. If it can be done with the designer, how would it be done?
With the experimentation I have done so far, I can only find ways for it to predict a single value, not a collection or values.
Thanks,
Simon
Aucun commentaire:
Enregistrer un commentaire