Deployment and Ensemble Machine Learning RRS feed

  • Question

  • So I have made a regression model that uses stacking of multiple algorithm results into one final model. 

    The issue is that when I go to test the model that I have deployed through Azure ML,  the input schema expects the intermediate results as inputs into the model. The problem is that I don't know the intermediate data because it is produced through the process of stacking the models. I wont know it before hand.

    This is similiar to getting rid of scored labels from the input schema when you test results except it is not as easy because the intermediate data (results from other algorithms) are essential to the prediction and scoring of it. Any idea how to navigate this issue? 

    Friday, August 23, 2019 12:49 AM

All replies