import glob, os
from datetime import date
import warnings
warnings.filterwarnings('ignore')
# Plotly offline
from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot
init_notebook_mode(connected=True)
# Import lecture/stockage données
import numpy as np
import pandas as pd
# Visualisation
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.express as px
import plotly.graph_objs as go
import plotly.figure_factory as ff
import calmap
import folium
from IPython.core.display import display, HTML
# Palette
colorGrey = '#393e46'
colorRed = '#ff2e63'
colorCyan = '#21bf73'
colorYellow = '#fe9801'
# On charge et map l'ensemble des csv en un seul
# dfGlobalDaily = pd.concat(map(pd.read_csv, glob.glob(os.path.join('', "../COVID-19/csse_covid_19_data/csse_covid_19_daily_reports/*.csv"))))
# dfGlobalDailyLatest = pd.read_csv("../COVID-19/csse_covid_19_data/csse_covid_19_time_series/time_series_19-covid-Confirmed/" + str(d1) + ".csv")
dfGlobalDailyLatest = pd.read_csv("../COVID-19/csse_covid_19_data/csse_covid_19_daily_reports/03-23-2020.csv")
# Chargement des times series pour les cas confirmés
dfGlobalSeriesConfirmed = pd.read_csv("../COVID-19/csse_covid_19_data/csse_covid_19_time_series/time_series_19-covid-Confirmed.csv")
# Chargement des times series pour les cas décédés
dfGlobalSeriesDeaths = pd.read_csv("../COVID-19/csse_covid_19_data/csse_covid_19_time_series/time_series_19-covid-Deaths.csv")
# Chargement des times series pour les cas soignés
dfGlobalSeriesRecovered = pd.read_csv("../COVID-19/csse_covid_19_data/csse_covid_19_time_series/time_series_19-covid-Recovered.csv")
dfSpecialFranceRaw = pd.read_csv('D:\\DeeplyLearning\\Github\\data\\dist\\chiffres-cles.csv')
dfSpecialFranceRawDepartement = dfSpecialFranceRaw[(dfSpecialFranceRaw.granularite == 'departement') & (dfSpecialFranceRaw.date == '2020-03-23')]
dfSpecialFranceRawRegion = dfSpecialFranceRaw[(dfSpecialFranceRaw.granularite == 'region') & (dfSpecialFranceRaw.date == '2020-03-23')]
# On récupère que le row FRANCE depuis les times series des cas confirmés
frSeriesConfirmed = dfGlobalSeriesConfirmed[(dfGlobalSeriesConfirmed['Country/Region'] == 'France') & (dfGlobalSeriesConfirmed['Province/State'] == 'France')]
frSeriesConfirmed = frSeriesConfirmed.drop(columns=['Lat', 'Long'])
# On récupère que le row FRANCE depuis les times series des cas décédés
frSeriesDeaths = dfGlobalSeriesDeaths[(dfGlobalSeriesDeaths['Country/Region'] == 'France') & (dfGlobalSeriesDeaths['Province/State'] == 'France')]
frSeriesDeaths = frSeriesDeaths.drop(columns=['Lat', 'Long'])
# On récupère que le row FRANCE depuis les times series des cas soignés
frSeriesRecovered = dfGlobalSeriesRecovered[(dfGlobalSeriesRecovered['Country/Region'] == 'France') & (dfGlobalSeriesRecovered['Province/State'] == 'France')]
frSeriesRecovered = frSeriesRecovered.drop(columns=['Lat', 'Long'])
# On rassemble l'ensemble de nos données précédents
frSeriesAll = pd.concat([frSeriesConfirmed, frSeriesDeaths, frSeriesRecovered])
frSeriesAll = frSeriesAll.drop(columns=['Province/State', 'Country/Region'])
frSeriesAll = frSeriesAll.reset_index()
frSeriesAll = frSeriesAll.drop(columns=['index'])
# On transpose notre matrice pour avoir une colonne date, une colone décédés, une colonne soignés, une colonne soignés
frSeriesAll = frSeriesAll.transpose()
frSeriesAll = frSeriesAll.rename(columns={0: 'Confirmed', 1: 'Deaths', 2: 'Recovered'})
frSeriesAll = frSeriesAll.reset_index()
frSeriesAll = frSeriesAll.rename(columns={'index' : 'Date'})
figFrSeriesAll = frSeriesAll.melt(id_vars="Date",
value_vars=['Confirmed', 'Deaths', 'Recovered'],
value_name='Nombre de cas',
var_name='type')
iplot(px.bar(figFrSeriesAll,
x="Date",
y="Nombre de cas",
color='type',
title='France',
color_discrete_sequence=[colorYellow, colorRed, colorCyan],
barmode='group'))
# On rassemble par Date
temp = frSeriesAll.reset_index()
temp = temp.groupby(['Date'])['Confirmed', 'Deaths', 'Recovered'].sum().reset_index()
# Ajout colonne taux mortalité
temp['Mortality Rate'] = round((temp['Deaths']/temp['Confirmed']), 3)*100
temp = temp.fillna({'Mortality Rate': 0.00})
# Ajout colonne taux de guerison
temp['Recovery Rate'] = round((temp['Recovered']/temp['Confirmed']), 3)*100
temp = temp.fillna({'Recovery Rate': 0.00})
# On converti notre string en Date
temp['Date'] = pd.to_datetime(temp['Date'], format='%m/%d/%y')
# On trie en fonction de la date
temp = temp.sort_values(by='Date', ascending=True)
# On transforme les données en 2 colonnes pour afficher à notre schéma
temp = temp.melt(id_vars='Date', value_vars=['Recovery Rate', 'Mortality Rate'],
var_name='Ratio', value_name='Value')
iplot(px.line(temp, x="Date", y="Value", color='Ratio', log_y=True,
title='Taux de guerison & de létalité', color_discrete_sequence=[colorCyan, colorRed]))
speFranceTopReg = dfSpecialFranceRawRegion.groupby(['maille_nom'])['cas_confirmes', 'deces', 'reanimation'].sum().reset_index()
speFranceTopReg = speFranceTopReg.rename(columns={'maille_nom': 'Region', 'cas_confirmes': 'Cas confirme'})
fig = px.bar(speFranceTopReg.sort_values('Cas confirme', ascending=False).head(10).sort_values('Cas confirme', ascending=True),
x="Cas confirme", y="Region", text='Cas confirme', orientation='h',
width=800, height=500, range_x = [0, max(speFranceTopReg['Cas confirme'])+500])
fig.update_traces(marker_color=colorRed, opacity=0.6, textposition='outside')
iplot(fig)
speFranceTopDef = dfSpecialFranceRawDepartement.groupby(['maille_nom'])['cas_confirmes', 'deces', 'reanimation'].sum().reset_index()
speFranceTopDef = speFranceTopDef.rename(columns={'maille_nom': 'Departement', 'cas_confirmes': 'Cas confirme'})
fig = px.bar(speFranceTopDef.sort_values('Cas confirme', ascending=False).head(10).sort_values('Cas confirme', ascending=True),
x="Cas confirme", y="Departement", text='Cas confirme', orientation='h',
width=800, height=500, range_x = [0, max(speFranceTopDef['Cas confirme'])+50])
fig.update_traces(marker_color=colorRed, opacity=0.6, textposition='outside')
iplot(fig)
source data : https://github.com/CSSEGISandData/COVID-19