Analyzing the composition of beer recipes and visualize results in a human-friendly way.
Check out the live website: https://www.beer-analytics.com/
Beer Analytics is a database of beer brewing recipes, built specifically for data analysis. It is made for beer enthusiasts and (home)brewers to provide detailed insights into brewing recipes, even when they're not an expert in data analysis. The goal is to expand the knowledge how certain types of beer are typically brewed, ultimately helping (home)brewers to compose better recipes themselves, and potentially uncover some trends in craft/home brewing.
The project has two main components:
- a recipe database with (hopefully) reliable data (clean and normalized, reduce outliers and bad data)
- a user interface to execute data analysis (filtering, slicing and dicing) and to present results in a visually appealing way
- Docker installed locally
- yarn (JavaScript package manager) installed locally
- Install yarn dependencies:
yarn install
- Create a configuration file (see below)
- Build and start the Docker container
docker compose up
- Jump into Docker container
docker exec -it beer_analytics_django bash
- Load initial data (known styles and ingredients) via
python manage.py load_initial_data
Provide a .env
file in the main folder. An example can be found in .env.example
.
Per default the application starts with "dev" settings, which is likely what you want. Use the DJANGO_SETTINGS_MODULE
environment variable to use different settings according to the environment:
# Dev settings
DJANGO_SETTINGS_MODULE=config.settings_dev
# Production settings
DJANGO_SETTINGS_MODULE=config.settings_prod
The Docker container uses dev settings.
To start the application for development run the Docker container
docker compose up
which starts a webserver at localhost:8000
.
In a second terminal run
yarn start
to start the Webpack dev server to compile CSS and JS files.
For legal reasons the project does not come with any recipe data included. You have to retrieve and import recipe data from the sources you'd like to analyze.
ℹ️ It is planned to add a database with anonymized data samples at some point. Sorry for inconvenience.
Recipes can be imported via CLI in various formats. Each recipe must have a unique id assigned, which can be an arbitrary string. The following recipe formats are supported with their respective commands:
python manage.py load_beerxml_recipe recipe.xml unique_id
python manage.py load_mmum_recipe recipe.json unique_id
python manage.py load_beersmith_recipe recipe.bsmx unique_id
Once recipes are imported, they need to be mapped to the list of known styles and ingredients. Run the following commands to execute the mapping. Any unmapped recipes will be processed:
python manage.py map_styles
python manage.py map_hops
python manage.py map_fermentables
python manage.py map_yeasts
These commands can be repeated any time and will process any recipes, which haven't been mapped yet. Please note that, depending on the amount if recipes, this step can take a while.
The application is pre-calculating and persisting some metrics for style and ingredients. To update these metrics, run:
python manage.py calculate_metrics
python manage.py calculate_hop_pairings
For information about the security policy and know security issues, see SECURITY.md.
This software is available under the GPLv3 license.
You're welcome to contribute new features, such as new analysis/chart types or bug fixes, by creating a Pull Request.
Please see CONTRIBUTING.md for more details.
Thank you @kasperg3 for sharing data from his awesome hops database.
I love to hear from people using my work, it's giving me the motivation to keep working on it.
If you want to let me know you're finding it useful, please consider giving it a star ⭐ on GitHub.
If you love my work and want to say thank you, you can help me out for a beer 🍻️ via PayPal.