Sorry, I'm no longer maintaining this project.
This software project tries to predict US COVID-19 cases and deaths.
Based on top of work by Anaiy Somalwar: https://github.com/anaiy2004/COVID-19-Forecasting
In my tests, the predicted deaths were better than predicted cases.
While this software project predicts COVID-19, deaths and especially cases are still random. USE THIS SOFTWARE AND DATA AT YOUR OWN RISK!
You need Python 3(at least 3.8 is recommended) and pip.
First, let's run pip install:
pip3 install -r requirements.txt
This might take some time.
Now, we need to download US data from JHU CSSE COVID-19 Data and sort the downloaded data by date:
python3 tools/download_us_data.py
python3 tools/date_sort.py US.csv US.csv
NOTE: The first argument passed to date_sort.py is the input file, US.csv and the second argument is the output file, again, US.csv
Finally let's predict deaths and cases:
python3 predict_deaths.py US.csv
python3 predict_cases.py US.csv
If you want to generate a data.json file, which contains predicted and current total cases and deaths in JSON, run:
python3 tools/create_data_json.py US.csv
The schema of data.json is in data.schema.json.
Having any issues? Simply open an issue.
Simply send a PR ;)
See LICENSE at the root.
There's also a data repo that contains prediction data in the form of data.json and dataset.csv files that are updated daily. See the README for more info.