A simple Python script (Jupyter Notebook : COVID19_Trend.ipynb) to forcast COVID-19 daily cases for a given country using Facebook Prophet and "Our World in Data" data source (continuously up-to-date).
Please set initial paramerters as required.
- Set data source : data_source = 'https://covid.ourworldindata.org/data/owid-covid-data.csv'
- Select Country : show_country = 'Spain'
- Adjusting trend flexibility : trend_flexibility = 1.0
- Set number of forcasting days : days_forcasting = 90
- Set number of previous days to be considered for forcasting : last_days_count = 30
- Facebook Prophet : "install -c conda-forge fbprophet"
- Plotly : "conda install -c plotly plotly"
- Pandas : "conda install -c conda-forge pandas"
- Matplotlib : "conda install -c conda-forge matplotlib"
- "Our World in Data" and its respective authors for providing up-to date data at https://github.com/owid/covid-19-data/tree/master/public/data
- Facebook, Inc. and its affiliates for "Facebook Prophet" library at https://github.com/facebook/prophet
- Authors and affiliates of all other libraries and technologies (including Python, Pandas, Plotly and Matplotlib) used in this script.
- Facebook Prophet : https://facebook.github.io/prophet/docs/quick_start.html#python-api
- "Our World in Data" Data Source CSV : https://github.com/owid/covid-19-data/blob/master/public/data/owid-covid-data.csv
- Trend Flexibility : https://facebook.github.io/prophet/docs/trend_changepoints.html#automatic-changepoint-detection-in-prophet
DataDisca Pty Ltd, Melbourne, Australia