The goal is to have a production ready code to analyse New York taxi data.

Ravioly is a way to encapsulate the raw data contained in NYC taxi csv files. As mentionned in the documentation (https://yleprince.github.io/ravioly/) it is built on top of pandas.DataFrame and provide specific processing functionalities and dedicated methods.

💡 Python ^3.7 is required.

Install using pip:

pip install git+https://github.com/yleprince/ravioly.git

In your python code:

>>> from ravioly.datastructure import Ravioly

>>> df = Ravioly('../data/nyc_data.csv', nrows=1000)
>>> df.km_by_dow()

day_of_week
0    480.647876
1    466.137703
2    553.287868
3    427.187865
4    465.982398
5    489.352866
6    557.113716
Name: km_by_dow, dtype: float64

• Dev: 💻

  • pandas: the code is a subclass of pandas.DataFrame data structure.
  • poetry: package management is made with poetry

• Lint: 📐

  • isort: to sort the imports
  • mypy: to check the use of types within the code
  • black: to format automatically the code with the pep8 requirements
  • flake8: to check code syntax

• Tests: 🧑‍🏫

  • pytest: to unit test the code
  • pytest-cov: to check percentage of code covered by the tests

• CI: 🤖

  • github actions: to maintain code consistency over commits.

• Documentation: 📚

  • sphinx: automatically generate the documentation from source code.
  • read the doc theme: theme for the documentation
  • github pages: to serve the documentation: https://yleprince.github.io/ravioly/

• Bonus: 🎁

  • pre-commit: pre-commit allows to run lint and tests workflow automatically at every step of the project
  • pre-push: pre-push allows to update the documentation every time the code is pushed on the github.