malteiwanicki / pynutriscore

pyNutriScore

The goal of the project is to provide simple package that can calculate the Nutri-Score.

What is the Nutri-Score ?

The Nutri-Score is a nutrition label that converts the nutritional value of products into a simple code consisting of 5 letters, each with its own colour.

Each product is awarded a score based on a scientific algorithm.

This formula takes into account the nutrients to avoid (energy value and the amount of sugars, saturated fats and salt) and the positive ones (the amount of fibre, protein, fruit, vegetables and nuts).

You can therefore see at a glance which products are recommended and which should be avoided.

Source: https://nutriscore.colruytgroup.com/colruytgroup/en/about-nutri-score/

Paper: https://pdfs.semanticscholar.org/3d1c/c206bc286bb5f80452821a0d26ff9e55b387.pdf

Where to get it

The source code is currently hosted on GitHub at: https://github.com/lemonhead94/pyNutriScore

Usage

Install

pip install pyNutriScore

Calculate the nutri-score:

from pyNutriScore import NutriScore

result = NutriScore().calculate(
    {
        'energy': 0,
        'fibers': 4,
        'fruit_percentage': 60,
        'proteins': 2,
        'saturated_fats': 2,
        'sodium': 500,
        'sugar': 10,
    },
    'solid'  # either 'solid' or 'beverage'
)

print(result)  # Output: 2

Calculate the nutri-score class:

from pyNutriScore import NutriScore

result = NutriScore().calculate_class(
    {
        'energy': 0,
        'fibers': 4,
        'fruit_percentage': 60,
        'proteins': 2,
        'saturated_fats': 2,
        'sodium': 500,
        'sugar': 10,
    },
    'solid'  # either 'solid' or 'beverage'
)

print(result)  # Output: "B"

License

MIT

Contribution

Feel free to add suggestions, PRs, comments and bug reports.

Credits

This is a re-implementation of the TypeScript package nutri-score by Alex Kolarski in Python.

Authors

Jorit Studer ([email protected])