Field note · December 12, 2023
Course note — Validation that matches deployment
Cross-validation is not one technique. Choosing the wrong variant is how models get shipped with an offline score that has no relationship to what happens next.
Revised Validation that matches deployment this week. The change was small and the reason was not.
Cross-validation is not one technique. Choosing the wrong variant is how models get shipped with an offline score that has no relationship to what happens next.
The first draft explained the mechanism and stopped. What was missing was the failure: the thing that happens when you get it wrong, described concretely enough that someone recognises it later. A lesson that only teaches the correct version leaves you unable to spot the incorrect one, which is the situation you will actually be in.
It sits in the Modelling, honestly module. The exercise is a thinking one rather than a query one — some of this material is about deciding what to compute, and that does not need a keyboard.
Ordering matters here more than in most courses. It follows Frame the problem, then beat a stupid baseline and leads into Features, and explaining what the model did, and reading it out of sequence mostly works but costs you the setup.
Free means free, and it also means we can rewrite it whenever it is wrong. No edition, no errata PDF, nothing to repurchase. The whole course is 36 lessons and the fixes land the day we find them.