The art and science of making models that actually work
A rigorous, story-driven journey through hyperparameter optimization — from understanding what hyperparameters are and why they matter, through the mathematics of search, all the way to AutoML, experiment tracking, and production best practices.
16 chapters across 5 parts — the full book for this course.
No previous year questions have been added yet.
3 chapters · 53 min
2 chapters · 58 min
3 chapters · 91 min
5 chapters · 112 min
Real pipelines, real datasets, real decisions
Logging runs so you can actually learn from them
Seeing the search landscape to understand what you found
Hard-won rules for running HPO that doesn't waste compute
What goes wrong, and exactly how to fix it
The knobs you turn before training even begins