Question

A data scientist fits a scikit-learn OneHotEncoder on a training set containing a categorical 'region' column with values {'North', 'South', 'East'}. During batch inference, the incoming data contains a new value 'West' that was never seen during training. With the encoder's default configuration, what happens, and what is the recommended fix so inference does not fail?

Accepted Answer

The default encoder raises an error on the unseen 'West' value; set handle_unknown='ignore' so unknown categories are encoded as all-zeros instead of failing.

Answer

The default encoder automatically creates a new column for 'West'; no change is needed because it adapts to unseen categories at transform time.

Answer

The default encoder silently drops the 'West' row; set handle_unknown='error' to keep those rows in the output.

Answer

The default encoder maps 'West' to the most frequent training category; set handle_unknown='ignore' to instead map it to the mean of the numeric columns.