Exercise

Iterative imputation

In the previous exercise, you derived mean imputations for missing values of loan_data. However, in a machine learning interview, you will probably be asked about more dynamic imputation techniques that rely on other features in the dataset.

In this exercise, you'll practice a machine-learning based approach by imputing missing values as a function of remaining features using IterativeImputer() from sklearn.impute. This is a multivariate imputer that estimates each feature from all of the others in a 'round-robin' fashion.

Note that this function is considered experimental, so please read the documentation for more information.

You're at the same place in the Pipeline:

Machine learning pipeline

Instructions 1/4

undefined XP
    1
    2
    3
    4
  • Subset loan_data's numeric columns and assign them to numeric_cols.