Skip to content

Project Structure

A generated project has the following layout:

my-project/
├── data/
│   ├── raw/                # Original, immutable data
│   ├── interim/            # Intermediate processed data
│   ├── processed/          # Final, analysis-ready datasets
│   └── external/           # Data from third-party sources
├── docs/                   # Documentation
├── models/                 # Saved models
├── notebooks/              # Jupyter notebooks
│   └── notebook_utils.py   # Utility for simplifying imports in notebooks
├── references/             # Data dictionaries, manuals, explanatory materials
├── reports/
│   ├── figures/            # Figures used in reports
│   ├── _output/            # Rendered output (gitignored)
│   └── _combined_template.{ipynb,qmd}  # Quarto report templates
├── src/
│   └── my_project/
│       ├── __init__.py     # Package initialization
│       ├── config.py       # Environment and path configuration
│       ├── dataset.py      # Data pipeline definition
│       ├── dataset_utils.py # Pipeline framework classes
│       ├── train.py        # Model training
│       └── predict.py      # Model inference
├── tests/                  # Test files
├── .env                    # Environment variables (gitignored)
├── .pre-commit-config.yaml # Pre-commit hook configuration
├── justfile                # Task runner commands
├── pyproject.toml          # Python project configuration
├── README.md               # Project README
└── README.template.md      # Template usage guide

Conventions

Notebook naming: <number>-<initials>-<description>.ipynb (e.g., 1.0-jqp-initial-data-exploration).

Editable install: The src/ package is installed in editable mode via pyproject.toml, so you can import my_project from notebooks and scripts without path hacking.