03_preprocessing -- sparse is sparse_output for newer versions of sklearn's OneHotEncoder #69

ISSUE: sparse is sparse_output for newer versions of sklearn's OneHotEncoder

LOCATION:

• 03_preprocessing notebook
• Code block in "Categorical Variable Encoding (One-hot & Dummy)" section

PROPOSED SOLUTION:
Replace sparse with sparse_output on this line =>

dummy_e = OneHotEncoder(categories='auto', drop='first', sparse=False)

Or, add a comment noting this potential change for newer sklearn versions. This error came up during a consulting session debugging issues a consultee encountered running the preprocessing notebook (https://github.com/dlab-consulting/requests/issues/2876).

RELATED REFERENCE: https://scikit-learn.org/dev/modules/generated/sklearn.preprocessing.OneHotEncoder.html