How to create an evaluation recipe with a non trained saved model (w/o active version)

Anis
Anis Dataiku DSS Core Designer, Dataiku DSS Adv Designer, Registered, Dataiku DSS Developer Posts: 18 ✭✭✭✭

Hi,

Is there any way to create an evaluation recipe in the flow against a saved model placeholder - a model version to be imported later from mlflow based training session through the api?

I get get an invalid argument error: No active version for model <saved_model.id>

Thanks

Answers

  • Anis
    Anis Dataiku DSS Core Designer, Dataiku DSS Adv Designer, Registered, Dataiku DSS Developer Posts: 18 ✭✭✭✭

    same question for scoring recipes. How can I use the Python API to create a scoring recipe for model that is not trained yet?

Setup Info
    Tags
      Help me…