/mdsf2018-prod

Production ready refactoring of the winning solution of the "meilleur data scientist de France" competition.

Primary LanguageHTML

mdsf2018-prod

This is a demonstration Code refactoring to enable models to go to production. It is based on the winner solution of the Meilleur data scientist de France competition.

Setup with conda

Setup your environment with conda:

conda env create -f env.yml

Activate the newly created conda environment:

conda activate mdsf2018

(Optional) Add your the conda environment as a jupyter kernel Your conda environment must be activated.

python -m ipykernel install --user --name mdsf2018 --display-name mdsf2018

Run

Command line:

python -m mdsf2018

Jupyter notebook/Jupyter lab:

  • Open the jupyter notebook
  • Set your kernel to mdsf2018
  • Enjoy

Configuration

If you want to submit a solution, you will have to sign up and get your token from the meilleur data scientist de France competition website.