A cookiecutter template for analytical, code-based projects within Her Majesty's Government.
- Who/what is this for?
- Getting started
- Changes to make post-creation
- Changes to consider post-creation
- Licence
- Contributing
- Acknowledgements
This template is for any HM Government analyst(s) who codes! It's main purpose is to:
- Provide a lightweight, Agile-like approach to Analytical Quality Assurance (AQA)
- Help quickly set up standardised project structures.
For more information about AQA, see The Aqua Book, and its resources.
Rather than using different files and locations to store AQA and documentation, our intention is to centralise as much of this as possible onto your Git repository hosting platform, e.g. GitHub or GitLab.
We use nudges, such as checklists in pull/merge requests, to minimise the burden on contributors and reviewers to complete AQA checks. This results in faster iterative development and deployment, whilst ensuring HM Government-wide standards on assurance are met.
We have also included pre-commit hooks to prevent accidental committing of secrets1, large data files, and Jupyter notebook outputs for security purposes.
⚠️ Only Unix-based systems (macOS, Linux, ...), and Python projects for GitHub or GitLab are supported — feel free to contribute to support other operating systems/programming languages!
To use this template to start your next coding project, make sure your system meets the requirements.
Once you're all set up, open your terminal, navigate to the directory where you want your new repository to exist, and run the following commands:
cookiecutter https://github.com/ukgovdatascience/govcookiecutter
Follow the prompts, and that's it — you've created your project structure! Post-creation, there are a few mandatory changes to make, as well as a some optional changes you should also consider making.
Otherwise, that's it — happy coding! 🎉
ℹ️ Contributors have some additional requirements! Check out the contributing guidelines for further details.
To get started your system should meet the following requirements:
- Unix-based system (macOS, Linux, ...)
- Python 3.5+ installed
- The
cookiecutter
package installed
There are many ways to install the cookiecutter
package. Our recommendation is to install it at the
system or user level, rather than as a Python package via pip
or conda
. This ensures it is isolated from the rest
of your system, and always available.
For macOS, open your terminal, and install cookiecutter
via Homebrew:
brew install cookiecutter
For Debian/Ubuntu, use the following commands:
sudo apt-get install cookiecutter
Otherwise, you can install cookiecutter
via pip
— you may wish to create a virtual environment first:
python3 -m pip install --user cookiecutter
Here's a few changes you must do once you've created your new project:
- Set up a Python virtual environment — there are many ways to set up a virtual environment, so we'll let you decide what's best for you!
- Git is not set up by default — open your terminal, navigate to your new project, run
git init
to set it up
Here's a few changes you should consider changing once you've created your new project:
- Make sure the
README.md
reflects what you want to do with your project! - Have a look inside the
docs/aqa
folder; you may want to modify some of them, e.g. the AQA plan - Want to add some project-specific checklists to the pull/merge request template? See the relevant Markdown files
within the
.github
(GitHub) or.gitlab/merge_request_templates
(GitLab) folder
Unless stated otherwise, the codebase is released under the MIT License. This covers both the codebase and any sample code in the documentation. The documentation is © Crown copyright and available under the terms of the Open Government 3.0 licence.
If you want to help us build, and improve govcookiecutter
, view our contributing guidelines.
This template is based off the DrivenData Cookiecutter Data Science project, especially around the
template data and src folder structures, and the make help
commands in the Makefiles.
[1]: Only secrets of specific patterns are detected by the pre-commit hooks. See here for further details.