Light cookiecutter template for starting data science projects. Minimal enough to stay out of your way when starting a project, sufficiently adaptable to remain useful as the project evolves.
I used cookiecutter-data-science and Kedro for a bit, personally grew to prefer starting projects with a clutter-free bare minimum project structure, and gradually expanding the project as needed. Thus this repo.
After installing the cookiecutter Python package, run cookiecutter gh:tnwei/cookiecutter-datascience-lite to create a new project repo. Example:
tnwei@rama:~/projects$ cookiecutter gh:tnwei/cookiecutter-datascience-lite
project_name [Project name]: Iris Species Classification
project_slug [iris-species-classification]: iris
New project directory created at /home/tnwei/projects/iris
A default .pre-commit-config.yaml is included in the project dir.
To use it, install `pre-commit`, `cd` into the project dir and run:
$ cd iris
$ git init
$ pre-commit install -t commit-msg -t pre-commit
Refer to https://pre-commit.com/ for more info.Created repo structure:
tnwei@rama:~/projects/iris$ tree
.
├── assets
├── data
├── iris
│ ├── data.py
│ └── utils.py
├── logs
├── output
└── README.md