Exercise

Ensemble models for extra votes

The LassoCV() model selected 22 out of 32 features. Not bad, but not a spectacular dimensionality reduction either. Let's use two more models to select the 10 features they consider most important using the Recursive Feature Eliminator (RFE).

The standardized training and test data has been pre-loaded for you as X_train, X_test, y_train, and y_test.

Instructions 1/4

undefined XP
    1
    2
    3
    4
  • Select 10 features with RFE on a GradientBoostingRegressor and drop 3 features on each step.