What Encoding Is and Is Not - Categorical Features, Exactly

Close the boundary: categorical encoding rewrites tokens, not model claims.

highlighted = computed this step

What categorical encoding does

It rewrites category tokens into explicit integer columns for a later method to read.

\text{tokens}\rightarrow\{0,1\}\text{ columns}

What it does not do

It does not train a model and does not prove accuracy, calibration, model quality, or probability truth.

\text{preprocessing only}

The honest boundary

This book pins one category list, one one-hot table, and one unknown-category policy. It is NOT training, NOT learning, NOT generalization, NOT accuracy, NOT calibration, NOT a guarantee of model quality, and NOT probability truth.

\text{category tokens become explicit indicator columns}