Repository navigation
FAQ
Is this one application or a collection of exercises?
A collection. Each tpX folder is self-contained with its own dependencies and, where applicable, its own test suite. There's no shared entry point across labs.
Why does CI only run tp2 and tp4's test suites?
Those are the two labs with real, hermetic pytest suites that don't need external infrastructure. tp3 needs a Jenkins server, tp5 needs Docker to run its full stack meaningfully, tp6 is a written spec with no code, and tp7 needs a full DVC/MLflow environment and a dataset download. CI instead runs a syntax/import check on tp1, tp5, and tp7 so broken code is still caught without requiring that infrastructure.
Is tp6 actually implemented?
No — it's a lab specification describing the Kubernetes exercise (minikube, kubectl, namespaces, pods) as reference material. There's no manifest or deployment code in the repository.
What does the tp7 pipeline actually train?
A model on the UCI Adult Income dataset, orchestrated as DVC stages (download → preprocess → split) with training and inference scripts tracked via MLflow. The logged run reached about 85% test accuracy.
Do I need Poetry for every lab?
No, only tp2 and tp4, which ship a pyproject.toml. tp5 uses a plain requirements.txt, and tp7 lists its dependencies directly in the README's install command.
Can I run the Jenkins or SonarQube integrations without setting up those servers?
You can still run the underlying pytest suites locally (poetry run pytest) — that's what CI does. Actually exercising the Jenkinsfile or getting a SonarQube quality report requires pointing a real Jenkins/SonarQube instance at the repo, which is outside the scope of what's checked into it.