AI engineering glossary · Feature engineering data prep
What is unit vector scaling?
Unit vector scaling (also called Row normalisation, Normalizer): Rescaling each row (sample) so that its vector has length 1, usually using the L2 norm. It keeps the direction of the row but discards its magnitude, which is useful when only the relative pattern matters, as in cosine-similarity comparisons of text vectors.
Unit vector scaling: an example
A row [3, 4] has L2 length 5, so it becomes [0.6, 0.8] after row-wise normalisation.
Related terms
More Feature engineering data prep terms
- Unknown category handling
- Variance threshold
- Weight of evidence
- Window feature
- Winsorization
- Wrapper method
- Yeo-Johnson transform
- ADASYN
Where you learn it
Unit vector scaling is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.