This is a dashboard for Covid-19 case data, using python and a Jupyter notebook to analyse and display the data from Johns Hopkins University.
Analysis is done using Altair visualisation library, and automatically saved to powerpoint file.
The following were the steps used for first time setup on Mac.
## first time setup
cd <directory path>
conda create -n py3 python=3.7
conda activate py3
# jupyter notebook
conda install nb_conda
# altair for data visualisation
pip install altair vega_datasets
# dependencies for notebook
pip install matplotlib seaborn datetime
pip install python-pptx
pip install pycountry_convert
# Altair saver to save charts as files for slide creation
# https://github.com/altair-viz/altair_saver
pip install altair_saver
# altair_saver has dependencies which aren't so straightforward, at least on Mac
# Install selenium using the chromedriver package from: https://chromedriver.chromium.org
# choose appropriate zip from files ending "linx64" or "mac64" or "mac_arm64" or "win32"
chromedriver_latest_stable_release='106.0.5249.61'
chromedriver_name=chromedriver_mac_arm64
wget https://chromedriver.storage.googleapis.com/${chromedriver_latest_stable_release}/${chromedriver_name}.zip
unzip ${chromedriver_name}.zip
rm ${chromedriver_name}.zip
mv chromedriver /usr/local/bin/
# load jupyter notebook
# note, might need to ensure Jupyter notebook has correct python kernel selected first time
jupyter notebook covid.ipynb
After first-time setup, subsequent runs can simply use the following:
cd <directory path>
conda activate py3
# load jupyter notebook
jupyter notebook covid.ipynb