ML 9.4Lesson outlineLesson 4 of 5 in Module 9
Gradient Boosting in Practice
Configure a modern gradient-boosting library well: a loss that matches the target, histogram splits, native categories and gaps, early stopping, a tuning order, and constraints.
Video in production.
Lesson outline: Gradient Boosting in Practice
In this lesson12 parts
- 01A price for every policy
- 02The policies
- 03The loss that matches the target
- 04Histograms
- 05The leaf value is a Newton step
- 06Four libraries, one dataset
- 07Categories and gaps, natively
- 08The learning rate against the number of trees
- 09A model that must move one way
- 10Where it misleads, and the sealed policies
- 11Code
- 12Recap, challenge, and what comes next
Builds on
This lesson stands on its own. These go deeper into what it uses.