Visualize Coronavirus Cases Geographically in Python

Introduction

In this article, we will visualize coronavirus (COVID-19) cases geographically using Python.

Steps

Install the chart_studio in your environment. Now Plotly is a part of chart_studio

pip install chart_studio

The data is picked from here. * I am using it for demo purposes.

Import all the necessary libraries and setup your jupyter notebook for offline plotly usage

import pandas as pd
import chart_studio.plotly as py
import plotly.graph_objs as go
from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot
init_notebook_mode(connected=True)
# Lets get data from that url
df = pd.read_html('https://www.worldometers.info/coronavirus/#countries')
# data is imported as a list
# use the indexing to get your dataframe
df = df[0]
type(df)
df.info()

You will find lot of NA values. Let's clean up the data and also rename columns and change datatype

df.rename(columns = {"Country,Other": "COUNTRY"}, inplace=True)
# fill NaN and Change columntype
df["NewCases"] = df["NewCases"].fillna(0).astype('int')
df["TotalDeaths"] = df["TotalDeaths"].fillna(0).astype('int')
df["NewDeaths"] =df["NewDeaths"].fillna(0).astype('int')
df["TotalRecovered"] = df["TotalRecovered"].fillna(0).astype('int')
df["ActiveCases"] = df["ActiveCases"].fillna(0).astype('int')
df["Serious,Critical"] =df["Serious,Critical"].fillna(0).astype('int')
df.info()

Now, we will create a new column text to get all this information as a string in single column so that when you hover over a country, it's easy to display.

df['text'] = df.apply(lambda r : "Deaths: " + str(r.TotalDeaths + r.NewDeaths) + " Suspected: " + " " + str(r.NewCases + r.ActiveCases),
axis = 1)

Now, when you see the head of this dataframe, you will get a data like below.

	COUNTRY	TotalCases	NewCases	TotalDeaths	NewDeaths	ActiveCases	TotalRecovered	Serious,Critical	text
0	China	79252	428	2835	47	37323	39094	7664	Deaths: 2882 Suspected: 37751
1	S. Korea	2931	594	17	1	2890	24	7	Deaths: 18 Suspected: 3484
2	Italy	889	0	21	0	822	46	64	Deaths: 21 Suspected: 822
3	Diamond Princess	705	0	6	0	689	10	36	Deaths: 6 Suspected: 689
4	Iran	388	0	34	0	281	73	0	Deaths: 34 Suspected: 281

Now lets replace the countries with their respective codes. This countrycode.csv is attached in the code zip file.

country_code=pd.read_csv("countrycode.csv")
# Convert the dataframe to dictionary
country_code.set_index('COUNTRY', inplace=True)
dict_country_code = country_code.to_dict()
REPLACE_LIST = dict_country_code['CODE']
# Replace Country with Codes
df.replace(REPLACE_LIST, inplace=True)

Now we need to begin to build our data dictionary. The easiest way to do this is to use the dict() function of the general form:

type = 'choropleth',
locations = country code
colorscale= Either a predefined string: 'pairs' | 'Greys' | 'Greens' | 'Bluered' | 'Hot' | 'Picnic' | 'Portland' | 'Jet' | 'RdBu' | 'Blackbody'
| 'Earth' | 'Electric' | 'YIOrRd' | 'YIGnBu'
or create a custom colorscale
text= list or array of text to display per point
z= array of values on z axis (color of state)
colorbar = {'title':'Colorbar Title'})
data = dict(
type = 'choropleth',
colorscale = 'ylorrd',
locations = df['COUNTRY'].values,
z = df['TotalCases'],
text = df['text'],
colorbar = {'title' : 'Corona Total Cases'},
)

Then we create the layout nested dictionary and Then we use

go.Figure(data = [data],layout = layout)

To set up the object that finally gets passed into iplot()

layout = dict(
title = 'Global Corona stats',
geo = dict(
showframe = False,
projection = {'type':'natural earth'}
)
)
choromap = go.Figure(data = [data],layout = layout)
iplot(choromap)

You will see the geographical map as we aimed for.

When you hover over each country, you will see data as in our text column.

Along with this, plotly has commands to hover, zoom in, zoom out, and save pictures.

From the data, only the Diamond Princess data is not shown.

The 'type' property is an enumeration that may be specified as ['equirectangular', 'mercator', 'orthographic', 'natural earth', 'kavrayskiy7', 'miller', 'robinson', 'eckert4', 'azimuthal equal area', 'azimuthal equidistant', 'conic equal area', 'conic conformal', 'conic equidistant', 'gnomonic', 'stereographic', 'mollweide', 'hammer', 'transverse mercator', 'albers usa', 'winkel tripel', 'aitoff', 'sinusoidal']

With any other data that is across the globe, you can create geographical charts. Plotly is just one of the ways, but it can be done in many other ways.