Glossary

Every concept in this sheet, plus the smaller terms it uses. Select a concept to open its card.

Config and secrets: Settings that change between runs or environments, such as paths, hyperparameters, model names and API keys, kept out of the code: parameters in a versioned config file, secrets in environment variables or a secrets manager.
Container image: A packaged file system with your code, a Python runtime and locked dependencies, built from a Dockerfile, that runs the same way on a laptop, in CI and in the cloud.

Data validation: Automatic checks on data before it is used: expected columns and types, value ranges, share of missing values, uniqueness of keys and row counts within a band.
Dependencies and lockfiles: A lockfile records the exact version of every package, including indirect ones, that your project resolved to. uv, a fast Python package and project manager, writes uv.lock and recreates the same environment anywhere with one command.
DVC (Data Version Control): An open source command-line tool that versions large files next to Git.
DVC pipelines: Stages declared in dvc.yaml, each with a command, its dependencies and its outputs.

Frozen evaluation set: A fixed, versioned set of examples with known correct answers that every candidate model or prompt is scored on, kept separate from training data and changed only deliberately.

Lineage: The recorded chain from a prediction back to the model version, the training run, the code commit, the parameters and the data version that produced it.

Model artefact formats: The file a trained model is saved as: joblib or pickle for scikit-learn objects, skops for a safer scikit-learn format, the framework's own format for PyTorch or XGBoost, and ONNX as a portable format many runtimes can load.

Project layout and pyproject.toml: A repository where notebook code becomes importable modules under src/<package>/, with dependencies, entry points and tool settings declared in one pyproject.toml file.

Tests for ML code: Fast tests with pytest for the deterministic parts: feature functions, data loading, schemas and output shapes, plus a smoke test that trains on a tiny sample and checks predictions have the right type and range.
The MLOps lifecycle: The loop a production model lives in: data is collected and validated, a model is trained and evaluated, approved and deployed, served to users, monitored, and retrained when it decays.
Tool map per stage: A reference of common open source tools for each step of the lifecycle.
Training entry point: One command that trains the model from start to finish: it reads config, loads data, fixes random seeds, fits the pipeline and saves the artefact, with no manual steps in between.