Fit your first model (Optional)

Holding data back (Optional)

Train a classifier and score it (Optional)

Scaling, and why a pipeline (Optional)