Get standarization parameters from Dataiku models

Options
rtruan
rtruan Dataiku DSS Core Concepts, Registered, Frontrunner 2022 Finalist, Frontrunner 2022 Participant Posts: 8 ✭✭✭✭
edited July 16 in Using Dataiku

Hello,

I understand that Dataiku is using for scaling the sklearn.preprocessing package. How can I retrieve the scale used by Dataiku in a particular model the same way i can do in sklearn?:

scaler = preprocessing.StandardScaler().fit(X_train)
 scaler.scale_

I cannot find in the documentation the way to get this data using the Dataiku API.

Best regards

Answers

  • MehdiH
    MehdiH Dataiker, Dataiku DSS Core Designer, Dataiku DSS Core Concepts Posts: 21 Dataiker
    edited July 17
    Options

    Hi @rtruan

    Through the Dataiku API, you can access the preprocessing settings used for a given feature for a prediction model with:

    dataikuapi.dss.ml.DSSTrainedPredictionModelDetails.get_preprocessing_settings()

    However, as of today, you cannot directly access the rescaling results through the API.

    Cheers

Setup Info
    Tags
      Help me…