My custom python data science cookiecutter for a poetry and pyenv user, which includes:
- pre-commit hooks:
autopep8poetry exportto sync dependencies with a requirements.txt file
- A local
.python-versionforpyenv - A
dataFolder structure similar tokedro - Optional dependencies groups for
plotlypoetry install --with plotly
- Optional dependency group for
rapidsassuming you have cuda 12 installed and are using an NVIDIA gpupoetry install --with rapids
polarsincluded
To create a new project folder that follows this cookiecutter template run:
python -m cookiecutter [email protected]:banditkings/ds_cookie_cu12.git
or if using HTTPS auth:
python -m cookiecutter https://github.com/banditkings/ds_cookie_cu12.git
Then you can navigate to the folder inside and install the library with required dependencies using:
poetry install --with dev,plotly,rapids{{ cookiecutter.repo_name }} <- Git repo name
├── .gitignore <- Hidden file that prevents staging of unnecessary files to `git`
├── README.md <- The top-level README for developers using this project
│
├── data <- Store raw data, features, etc - not committed to git
│ ├── 01_raw <-- Raw immutable data
│ ├── 02_intermediate <-- Typed data
│ ├── 03_primary <-- Domain model data
│ ├── 04_feature <-- Model features
│ ├── 05_model_input <-- Data ready for model input
│ ├── 06_models <-- Serialised models
│ ├── 07_model_output <-- Data generated by model runs
│ └── 08_reporting <-- Ad hoc descriptive cuts
│
├── notebooks <- Jupyter notebooks. Naming convention is a number (for ordering),
│ the creator's initials, and a short `-` delimited description, e.g.
│ `1.0-jqp-initial-data-exploration`.
│
├── .pre-commit-config.yaml <- git pre-commit config file
│
├── .python-version <- Python version number for local `pyenv` environment
│
├── Makefile <- Makefile to quickly rerun data pipelines, etc
│
├── pyproject.toml <- Poetry dependency and environment file
│
├── reports <- Generated analysis as HTML, PDF, LaTeX, etc
│ └── figures <- Generated graphics and figures to be used in reporting
│
└── {{ cookiecutter.package_name }} <- Source code for this project
├── tests <- All tests for this package
├── data <- Functions to ETL data and perform feature engineering
├── evaluation <- Custom metrics and evaluation criteria
├── experiments <- Experiment tracking and logging logic
├── modeling <- Helper classes to train models and use trained models to make predictions
└── utils <- Generic Utility functions
The project starts off with pre-commit hooks with some autopep8 rules and trailing whitespace checks, followed by some poetry lock fixes that can take a long time to resolve dependencies.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent