Skip to content

EX-07 · Chunking bake-off on your own corpus

Difficulty: 🟡 · label difficulty: 2 Stuck? Ask in the clinic thread for EX-07 — one long-running thread per exercise, so the answers accumulate where the next person will look. Submit: post it in Show and Tell titled [solution · EX-07] ..., with your numbers and your interval.

Notebook 03 · ~2 h · Skill: choosing a strategy from corpus shape

Run all seven strategies against a corpus of your own (or an unbalanced variant of this one). Report recall, storage multiplier, index cost, and unresolvable gold spans.

Acceptance criteria

  • The seven-row table
  • The corpus profile that drove the decision tree, derived from measurement
  • A one-paragraph recommendation naming what it costs