Plotly 활용한 covid-19 데이터 시각화

covid-19 데이터를 시각화로 훑어보자

1. 기간 : 2020.01.22~2020.03.22

2. data : 공공데이터를 bigquery에 로드하여 분석

3. 시각화 툴 : plotly express 를 주로 사용

import numpy as np
import pandas as pd

import chart_studio.plotly as py
import plotly.express as px
import plotly.graph_objects as go

import cufflinks as cf
cf.go_offline(connected=True)

%matplotlib inline
%load_ext google.cloud.bigquery

 

%%bigquery df
with
data as (
SELECT
  _TABLE_SUFFIX as dt,
  *
FROM 
  `data.coronavirus.csse_covid19_*` 
WHERE 
  _TABLE_SUFFIX BETWEEN '20200122' AND '20200322'

)
SELECT * FROM data 

df.head()

Out[3]:

	dt	Province_State	Country_Region	LastUpdate	Confirmed	Deaths	Recovered	Latitude	Longitude
0	20200211	Zhejiang	Mainland China	2020-02-11T12:53:02	1117	0	270	0.0	0.0
1	20200211	Jiangsu	Mainland China	2020-02-11T08:13:06	515	0	93	0.0	0.0
2	20200211	Fujian	Mainland China	2020-02-11T14:03:05	267	0	45	0.0	0.0
3	20200211	Shaanxi	Mainland China	2020-02-11T06:03:16	219	0	32	0.0	0.0
4	20200211	Yunnan	Mainland China	2020-02-11T09:23:04	153	0	20	0.0	0.0

국가별 확진자수

df_nation_confirmed = df.groupby(['Country_Region']).sum()[['Confirmed','Deaths']]
df_nation_confirmed = df_nation_confirmed.reset_index()
df_nation_confirmed.head()

 

	Country_Region	Confirmed	Deaths
0	Azerbaijan	1	0
1	Afghanistan	249	1
2	Albania	624	17
3	Algeria	991	77
4	Andorra	423	1

fig = px.bar(df_nation_confirmed, x='Country_Region', y="Confirmed")
fig.show()

 

 

국가별 사망자수

fig = px.bar(df_nation_confirmed, x='Country_Region', y="Deaths")
fig.show()

 

국가별 확진자 수, 사망자수 현황

fig = px.scatter_mapbox(df, lat='Latitude', lon='Longitude', hover_name='Country_Region', hover_data=['Confirmed','Deaths'],
                        color_discrete_sequence=["fuchsia"], zoom=2, height=600)
fig.update_layout(mapbox_style="open-street-map")
fig.update_layout(margin={"r":0,"t":0,"l":0,"b":0})
fig.show()

 

 

한국의 확진자 분포 확인

df_korea = df[df.Country_Region=='South Korea']
df_korea.head()

 

	dt	Province_State	Country_Region	LastUpdate	Confirmed	Deaths	Recovered	Latitude	Longitude
13	20200211	0	South Korea	2020-02-11T02:53:02	28	0	3	0.0	0.0
370	20200206	0	South Korea	2020-02-06T02:53:03	23	0	0	0.0	0.0
435	20200209	0	South Korea	2020-02-09T02:33:02	25	0	3	0.0	0.0
740	20200128	0	South Korea	1/28/20 23:00	4	0	0	0.0	0.0
1367	20200202	0	South Korea	2020-02-02T02:23:13	15	0	0	0.0	0.0

fig = px.box(df_korea, y="Confirmed")
fig.show()

 

 

일별 국가별 확진자 수

%%bigquery df_dt
with
data as (
SELECT
  _TABLE_SUFFIX as dt,
  *
FROM 
  `data.coronavirus.csse_covid19_*` 
WHERE 
  _TABLE_SUFFIX BETWEEN '20200122' AND '20200322'

)
SELECT PARSE_DATETIME('%Y%m%d',CAST(dt AS STRING)) AS dt,Country_Region, SUM(Confirmed) as Confirmed, Sum(Deaths) as Deaths
FROM data 
GROUP BY dt, Country_Region
ORDER BY Confirmed

 

df_dt.head()

 

	dt	Country_Region	Confirmed	Deaths
0	2020-01-23	Philippines	0	0
1	2020-03-16	occupied Palestinian territory	0	0
2	2020-03-15	occupied Palestinian territory	0	0
3	2020-03-22	Guernsey	0	0
4	2020-03-21	Republic of the Congo	0	0

fig = px.bar(df_dt, x='dt',y="Confirmed",hover_name='Country_Region',color='Country_Region')
fig.show()

 

bar plot을 이용하면 일별 국가별 누적 확진자처럼 한 날짜에 카테고리별 누적값을 추가할 수 있다.

fig = px.line_3d(df_dt, x='dt', y='Country_Region', z="Confirmed",hover_name='Country_Region',color='Country_Region')
fig.show()