An interactive notebook for examining trends in confirmed COVID-19 cases
This project is an attempt to track the COVID-19 pandemic in the United States. It uses data from the New York Times COVID-19 dataset and compiles that information into a series of choropleth maps with county-level detail. Clicking on a county triggers historical data to load, at which point forecasts of possible future transmission can be generated.
Yes! The notebook is currently not hosted on a permanent web site, though there are plans to do this; however, all the code is production-ready and can be run from the repository itself if you have Python installed. If you encounter an issue or have any suggestions, let us know by submitting an issue to this repository.
The notebook itself is simply a Dash app. Since this isn't in production yet, there is no permanent website where the results are hosted, but the program can be installed and run locally at any time.
First, clone the repository and create a virtual environment:
git clone https://github.com/epispot/covid-notebook
cd covid-notebook
conda create -n epinote # or, use pip if you prefer
Next, ensure that you are running Python 3.10 or later. Due to certain new syntax introduced in 3.10, the app and relevant scripts will not work on older runtimes. If you're using Anaconda, you can simply install the correct version of Python with:
conda install python==3.10.6
Finally, install all of the base requirements before starting the app.
This project uses Poetry for dependency management, but a requirements.txt
file is listed for pip
, though this won't install dev dependencies (use dev-requirements.txt
after for that):
poetry install
# pip install -r requirements.txt
# pip install -r dev-requirements.txt
You're all set! 🚀
To start a sample instance for debugging, simply run python app.py
.
Otherwise, use gunicorn
to create a production-ready server instance:
gunicron app:server