Recursive Feature Elimination with random forests
You'll wrap a Recursive Feature Eliminator around a random forest model to remove features step by step. This method is more conservative compared to selecting features after applying a single importance threshold. Since dropping one feature can influence the relative importances of the others.
You'll need these pre-loaded datasets: X
, X_train
, y_train
.
Functions and classes that have been pre-loaded for you are: RandomForestClassifier()
, RFE()
, train_test_split()
.
This exercise is part of the course
Dimensionality Reduction in Python
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Wrap the feature eliminator around the random forest model
rfe = ____(estimator=____, n_features_to_select=____, verbose=1)