Covid-19 Forecaster ©

App description

This is an interactive web application which offers tools for visualising and forecasting data related to Covid-19 in Germany, starting from early beginnings of the virus evolution.

The whole development process was based on OSEMN methodology: Obtain, Scrub and Explore data, Model and iNterpret the data, Deploy the model.

Dataset used: https://www.kaggle.com/headsortails/covid19-tracking-germany

The project includes the following modules:

• expl_analysis.ipynb the file that contains all the details about data processing and exploratory analysis
• forecasting_choice.ipynb the file which contains all information regarding the forecaster choise for prediction model
• app.py the main script which is responsible for model deployment

To run this project, you first should install the following dependencies by running the following command:

$ pip install -r requirements.txt

Make sure that you have changed the file paths cases_file_path, deaths_file_path and recovered_file_path in app.py to the actual path of the following files:

• cases_covid_de.csv
• deaths_covid_de.csv
• recovered_covid_de.csv

in data folder which you already cloned to your local machine.

If you want to update the csv files mentioned above with new data, run expl_analysis.ipynb using Google Colaboratory.

Then start app.py in folder code, by running this command in your terminal:

python app.py

Click on the URL generated by Streamlit and you should be redirected to the web page which should look like this: