Exercises¶
Page Maps¶
graph LR
family["Reproducible Research"]
program["Deep Dive DVC"]
section["Recovery Scale Incident Survival"]
page["Exercises"]
capstone["Capstone evidence"]
family --> program --> section --> page
page -.applies in.-> capstone
flowchart LR
orient["Orient on the page map"] --> read["Read the main claim and examples"]
read --> inspect["Inspect the related code, proof, or capstone surface"]
inspect --> verify["Run or review the verification path"]
verify --> apply["Apply the idea back to the module and capstone"]
Use these exercises to practice long-lived recovery judgment, not only DVC command vocabulary.
The strongest answers will name what must survive, where it is stored, how it is checked, and what can safely expire.
Exercise 1: Name the durability boundary¶
A release has:
publish/v1/manifest.jsonpublish/v1/metrics.jsonpublish/v1/params.yaml- DVC-tracked model output referenced in
dvc.lock
Write a short recovery goal that explains what must survive local cache loss and what a maintainer should run to verify it.
Exercise 2: Classify retention value¶
Classify these states as protected, bounded-retention, or safe-to-discard after review:
- promoted release artifact
- current mainline training data
- abandoned exploratory candidate output
- published analysis dataset
- temporary local debug report
Explain one sentence for each classification.
Exercise 3: Review a cleanup request¶
A teammate wants to run:
because storage is getting expensive.
Write a review response that explains what evidence you need before approving cleanup.
Exercise 4: Plan a remote migration check¶
A team is moving from an old DVC remote to a new one.
Describe a migration check that covers:
- inventory of important states
- copying or pushing required objects
- clean checkout verification
- rollback or cutover decision
Exercise 5: Write an incident note¶
CI started producing different metrics after a base image update.
Write an incident note that explains:
- what changed
- why the executor is part of the evidence story
- what should be checked before accepting or rejecting the new results
- what documentation or policy should be updated
Mastery check¶
You have a strong grasp of this module if your answers consistently keep five ideas visible:
- local cache is not durable authority
- retention follows value, not age alone
- cleanup needs reference scope and dry-run review
- remote migration and CI drift can break recovery
- incident response should preserve evidence before repair