Log, normalization and scaling
Why not use the raw numbers? Three problems, three fixes, and each fix has its own job: a log transform for huge and tiny values, normalization for samples measured more deeply, and scaling for big features that shout.
In this lesson
- What a log transform does to a skewed feature.
- Why samples need to be put on the same footing.
- When scaling features helps, and what it changes.
Narration is an AI-generated voice. Data shown are simulated. The video plays from YouTube in privacy-enhanced mode and loads only when you press play.