Awesome

2019-nCov noval coronavirus Data analysis in Python

简体中文 | English

Introduction on Medium

https://towardsdatascience.com/understanding-the-coronavirus-epidemic-data-44d2fb356ecb

Data Sources

Chinese Data:

• Raw Data from Ding Xiang Yuan.
• CSV format data from: https://github.com/BlankerL/DXY-2019-nCoV-Data CSV data file is updated frequently by 2019-nCoV Infection Data Realtime Crawler.

International Data:

• CSV daily data from John Hopkins: https://github.com/CSSEGISandData/COVID-19
• Real time JSON from https://github.com/ExpDev07/coronavirus-tracker-api , which also uses John Hopkins as underlying data.

Prerequisite

• Pandas
• If you need interactive analysis, and you cannot access Google Colab, then you need to install Python Notebook first.

Description

• coronavirus_demo_colab.ipynb: A demo on Google Colab, showing how to extract / aggregate / slice data, and basic time series / cross-sectional plotting
• demo.ipynb: Similar demo in a traditional Python Notebook, Chinese version
• demo.en.ipynb: Similar demo in a traditional Python Notebook, English version
• demo.html, demo.pdf: For those who doon't have Python Notebook, these two files serve as demo.ipynb for demonstration purpose (both are in Chinese)
• death_rate.ipynb: An example analysis of the heterogeneity of death rate across different regions
• utils.py: Utility functions

Some Examples:

data = utils.load_chinese_data()  # obtain CSV real time data
daily_frm = utils.aggDaily(data)  # aggregate to daily data
utils.tsplot_conf_dead_cured(daily_frm)  # Time Series plot of the Confirmed, dead, cured count of the whole country

Best Wishes