Questions on quick modeling prediction

YingFei
YingFei Registered Posts: 1

I have questions on the quick modeling part of dataiku. Now I am completing an assignment, but I find that the column of data in my label data used to calculate the cost does not appear in the unlabeled data. This problem caused me to be unable to predict unlabeled data with the model I trained. I would like to ask how to train the model and produce the corresponding decision rules without using the cost column to calculate the expected income? The cost and profit of the relevant decision rules are given in the question, but I cannot calculate them due to missing columns. Is there any other way to solve this problem?

Tagged:

Answers

  • Alexandru
    Alexandru Dataiker, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 1,250 Dataiker

    Hi Ying,

    Could you try to train your model using only the labeled features (excluding the cost column). If the cost column is not needed for model training, it is just for calculating expected income afterward.
    Use a Scoring Recipe in Dataiku to apply the trained model to your unlabeled dataset. This will generate predictions.

    Thanks

Setup Info
    Tags
      Help me…