Exercise

Loading models from the Model Registry

In this exercise, you will use the scikit-learn flavor to deploy the most stable "Insurance" model from the MLflow Model Registry and then use test data to get a prediction from the model.

The model uses LogisticRegression to predict whether an insurance claim is for a male or female, which is labeled as 1 or 0. You'll load the model and then make predictions using a test set called X_test.

The MLflow module will be imported.

Instructions

100 XP
  • With the scikit-learn flavor, load the "Production" version of the "Insurance" model using the convention for fetching models from the Registry as the model URI.
  • Using the loaded model, run a prediction on the test data from train_test_split used during training the model.