def show_balance_per_day(start, end, df):
""" Show the time series of the world: balance per day """
fig, ax = plt.subplots()
fig2, ax2 = plt.subplots()
start_index, end_index = get_start_end(start, end)
data = {}
variables = ['confirmed', 'recovered', 'deaths']
for variable in variables:
dates_, x = get_single_time_series(df, variable, start_index,
end_index)
data[variable] = cases_per_day(x)
data['resolved'] = data['recovered'] + data['deaths']
data['balance'] = data['confirmed'] - data['resolved']
for variable in ['confirmed', 'resolved', 'balance']:
ax.plot(dates_[1:],
data[variable],
'o-',
c=ws.varcolors[variable],
label=variable)
ax.plot(dates_[1:],
data['confirmed'],
'o-',
c=ws.varcolors['confirmed'],
label='confirmed')
ax.plot(dates_[1:],
data['resolved'],
'o-',
c=ws.varcolors['resolved'],
label='resolved')
# Date ticks
months = mdates.MonthLocator() # every month
days = mdates.DayLocator() # every month
month_fmt = mdates.DateFormatter('%m')
plt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%d/%m/%Y'))
utl.configure_axes(ax,
xlims=(dates_[1], dates_[-1]),
xlabel='Date',
ylabel='New Cases per Day')
ax.axhline(y=0, ls='--', c='gray')
utl.configure_axes(ax2,
xlims=(dates_[1], dates_[-1]),
xlabel='Date',
ylabel='New Cases per Day')
ax2.axhline(y=0, ls='--', c='gray')
ax2.set_ylim(0, data['confirmed'].max())
fig.savefig(os.path.join(ws.folders['website/static/images'],
'balance_per_day.png'),
bbox_inches='tight')
fig2.savefig(os.path.join(ws.folders['website/static/images'],
'balance_per_day_cleaner.png'),
bbox_inches='tight')
def show_cases_per_day(start, end, df):
""" Show the time series of the growth rate of the virus as new cases per day """
fig, ax = plt.subplots()
start_index, end_index = get_start_end(start, end)
data = {}
variables = ['confirmed', 'recovered', 'deaths']
for variable in variables:
dates_, x = get_single_time_series(df, variable, start_index,
end_index)
data[variable] = cases_per_day(x)
ax.plot(dates_[1:],
data[variable],
'o-',
c=ws.varcolors[variable],
label=variable)
# Second graph: only recovered and deaths, in log scale
fig2, ax2 = plt.subplots()
ax2.plot(dates_[1:],
data['recovered'],
'o-',
c=ws.varcolors['recovered'],
label='recovered')
ax2.plot(dates_[1:],
data['deaths'],
'o-',
c=ws.varcolors['deaths'],
label='deaths')
ax2.set_yscale('log')
utl.configure_axes(ax2,
xlims=(dates_[1], dates_[-1]),
xlabel='Date',
ylabel='New Cases per Day')
# Date ticks
months = mdates.MonthLocator() # every month
days = mdates.DayLocator() # every month
month_fmt = mdates.DateFormatter('%m')
plt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%d/%m/%Y'))
utl.configure_axes(ax,
xlims=(dates_[1], dates_[-1]),
xlabel='Date',
ylabel='New Cases per Day')
fig.savefig(os.path.join(ws.folders['website/static/images'],
'cases_per_day.png'),
bbox_inches='tight')
def show_world_death_ratio_I(start, end, df, country='world'):
""" Show the time series of the death ratios """
fig, ax = plt.subplots()
start_index, end_index = get_start_end(start, end)
data = {}
variables = ['confirmed', 'recovered', 'deaths']
for variable in variables:
dates_, data[variable] = get_single_time_series(
df, variable, start_index, end_index)
# Variable 'death ratio'
data['resolved'] = data['recovered'] + data['deaths']
data['deaths over total'] = data['deaths'] / data['confirmed']
data['deaths over resolved'] = data['deaths'] / data['resolved']
ax.plot(dates_,
data['deaths over total'],
c=ws.varcolors['deaths'],
ls='-',
label='deaths over total')
ax.plot(dates_,
data['deaths over resolved'],
c=ws.varcolors['deaths'],
ls='--',
label='deaths over closed cases')
# Date ticks
months = mdates.MonthLocator() # every month
days = mdates.DayLocator() # every month
month_fmt = mdates.DateFormatter('%m')
# format the ticks
if False:
ax.xaxis.set_major_locator(months)
ax.xaxis.set_major_formatter(month_fmt)
ax.xaxis.set_minor_locator(days)
plt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%d/%m/%Y'))
utl.configure_axes(ax,
xlims=(dates_[0], dates_[-1]),
xlabel='Date',
ylabel='Ratios')
fig.savefig(os.path.join(ws.folders['website/static/images'],
'world_death_ratio_I.png'),
bbox_inches='tight')
def show_country(country, variables, start, end, df, fig=None, ax=None):
""" Plot the time series of 'variable' from 'start' to 'end', and for 'country' """
if fig is None and ax is None:
fig, ax = plt.subplots()
start_index, end_index = get_start_end(start, end)
# Select the data for 'country'
df_ = df[df['Country/Region'] == country]
# 'variables' can be either one string or a list
if isinstance(variables, list):
pass
elif isinstance(variables, str):
# Turn it into a list
variables = [variables]
else:
print(
'ERROR: argument \'variables\' needs to be a string or list of strings'
)
return
for variable in variables:
data = []
dates_ = []
dates_, data = get_single_time_series(df_, variable, start_index,
end_index)
# Show time series
ax.plot(dates_, data, c=ws.varcolors[variable], label=variable)
utl.configure_axes(ax,
xlims=(dates_[0], dates_[-1]),
xlabel='Date',
ylabel='Cases')
ax.set_title(country)
# ax.set_yscale('log')
fig.savefig(os.path.join(
ws.folders['website/static/images'],
'%s_nolog.png' % country.replace(' ', '_').lower()),
bbox_inches='tight')
def show_world_time_series(start, end, df):
""" Show the time series of the world """
fig, ax = plt.subplots()
start_index, end_index = get_start_end(start, end)
data = {}
variables = ['confirmed', 'recovered', 'deaths']
for variable in variables:
dates_, data[variable] = get_single_time_series(
df, variable, start_index, end_index)
# Existing cases
data['resolved'] = data['recovered'] + data['deaths']
data['existing'] = data['confirmed'] - data['resolved']
variables = variables + ['existing']
for variable in variables:
ax.plot(dates_,
data[variable],
c=ws.varcolors[variable],
label=variable)
# Date ticks
months = mdates.MonthLocator() # every month
days = mdates.DayLocator() # every month
month_fmt = mdates.DateFormatter('%m')
# format the ticks
if False:
ax.xaxis.set_major_locator(months)
ax.xaxis.set_major_formatter(month_fmt)
ax.xaxis.set_minor_locator(days)
plt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%d/%m/%Y'))
utl.configure_axes(ax,
xlims=(dates_[0], dates_[-1]),
xlabel='Date',
ylabel='Cases')
ax.set_ylim(0, data['confirmed'].max())
fig.savefig(os.path.join(ws.folders['website/static/images'], 'world.png'),
bbox_inches='tight')
def countries_dayn(n, countries):
""" Show countries from the day they reached or surpassed n cases """
# Get data
df = ws.data_countries_only
# Parameters
linewidths = {k: 1 for k in countries}
linewidths['Colombia'] = 2
end = '01/04/2020'
# Create figures
fig1, ax1 = plt.subplots()
fig2, ax2 = plt.subplots()
fig3, ax3 = plt.subplots()
# Iterate over countries to plot data
for country in countries:
# Find the day when n cases were reached for each country
# First, get the whole time series
dates_, x = get_single_time_series(df[df['country_region'] == country],
'confirmed', 0,
ws.date_indices['01/04/2020'])
# Find where the country reached the n cases
date_index = np.where(x >= n)[0][0]
start = ws.dates_keys[date_index]
start_index, end_index = get_start_end(start, end)
data = {}
daysmin = 1e99
daysmax = -1e99
dates_, x = get_single_time_series(df[df['country_region'] == country],
'confirmed', start_index, end_index)
data['confirmed'] = x[:]
# Get new cases in 7 days
data_n7d = get_new_7_days(start_index,
end_index,
'confirmed',
country=country,
avg=True)
# Days
days = np.arange(len(dates_))
daysmin = min(days.min(), daysmin)
daysmax = max(days.max(), daysmax)
# CASOS CONFIRMADOS
ax1.plot(days,
data['confirmed'],
'o-',
lw=linewidths[country],
label=country)
# VELOCIDAD
c = data['confirmed']
velocidad = c[1:] - c[:-1]
velocidad = data_n7d['new_7_days']
# ax2.plot(days[1:], velocidad, 'o-', lw=linewidths[country], label=country)
ax2.plot(days, velocidad, 'o-', lw=linewidths[country], label=country)
# ACELERACIÓN
aceleracion = velocidad[1:] - velocidad[:-1]
# ax3.plot(days[2:], aceleracion, 'o-', lw=linewidths[country], label=country)
ax3.plot(days[1:],
aceleracion,
'o-',
lw=linewidths[country],
label=country)
# Arrange figures
# 1
ax1.set_ylim(0, 1100)
# ax1.set_yscale('log')
utl.configure_axes(ax1,
xlims=(daysmin, 14),
xlabel='Días después de alcanzar %i casos' % n,
ylabel='Casos confirmados')
ax1.set_title('Casos confirmados')
# 2
# ax2.set_ylim(0, 500)
ax2.set_ylim(0, 250)
# ax2.set_yscale('log')
utl.configure_axes(ax2,
xlims=(daysmin, 14),
xlabel='Días después de alcanzar %i casos' % n,
ylabel='Velocidad (casos confirmados nuevos diarios)')
utl.configure_axes(
ax2,
xlims=(daysmin, 14),
xlabel='Días después de alcanzar %i casos' % n,
ylabel=
'Velocidad (casos confirmados nuevos diarios;\npromedio de 7 días)')
ax2.set_title('Velocidad')
# 3
# ax3.set_ylim(-10, 100)
ax3.set_ylim(-10, 50)
# ax3.set_yscale('log')
ax3.axhline(y=0, ls='--', c='grey')
utl.configure_axes(
ax3,
xlims=(daysmin, 14),
xlabel='Días después de alcanzar %i casos' % n,
ylabel='Aceleración (aumento diario de casos confirmados nuevos)')
utl.configure_axes(
ax3,
xlims=(daysmin, 14),
xlabel='Días después de alcanzar %i casos' % n,
ylabel=
'Aceleración (aumento diario de casos confirmados nuevos;\npromedio de 7 días)'
)
ax3.set_title('Aceleración')
# Save figures
fig1.savefig(os.path.join(
ws.folders['website/static/images'],
'custom_casos_%i_%s_v2.png' % (n, ''.join([c[0] for c in countries]))),
bbox_inches='tight',
dpi=300)
fig2.savefig(os.path.join(
ws.folders['website/static/images'], 'custom_velocidad_%i_%s_v4.png' %
(n, ''.join([c[0] for c in countries]))),
bbox_inches='tight',
dpi=300)
fig3.savefig(os.path.join(
ws.folders['website/static/images'],
'custom_aceleracion_%i_%s_v2.png' %
(n, ''.join([c[0] for c in countries]))),
bbox_inches='tight',
dpi=300)
def analysis_03(start, end, df):
""" """
countries = ['Spain', 'Italy', 'Colombia']
countries = ['Spain', 'Italy', 'Colombia', 'Korea, South', 'France']
fig3, ax3 = plt.subplots()
colors = {
'Spain': 'k',
'Colombia': 'r',
'Italy': 'b',
'Korea, South': 'g',
'France': 'orange',
}
linewidths = {
'Spain': 1,
'Colombia': 2,
'Korea, South': 1,
'Italy': 1,
'France': 1,
}
start_by_country3 = {
'Spain': ws.dates_keys[33],
'Colombia': ws.dates_keys[47],
'Italy': ws.dates_keys[29],
'Korea, South': ws.dates_keys[28],
'France': ws.dates_keys[34],
}
for country in countries:
start = start_by_country3[country]
start_index, end_index = get_start_end(start, end)
data = {}
variables = ['confirmed', 'deaths', 'recovered']
daysmin = 1e99
daysmax = -1e99
for variable in variables:
dates_, x = get_single_time_series(
df[df['Country/Region'] == country], variable, start_index,
end_index)
data[variable] = x[:]
# Existing cases
data['resolved'] = data['recovered'] + data['deaths']
data['existing'] = data['confirmed'] - data['resolved']
days = np.arange(len(dates_))
daysmin = min(days.min(), daysmin)
daysmax = max(days.max(), daysmax)
ax3.plot(days,
data['existing'],
'o-',
lw=linewidths[country],
c=colors[country],
alpha=0.5,
label=country)
utl.configure_axes(ax3,
xlims=(daysmin, 15),
xlabel='Days after Growth Rate Departure',
ylabel='New Cases per Day')
ax3.set_ylim(0, 1000)
ax3.set_title('Existing Cases')
fig3.savefig(os.path.join(ws.folders['website/static/images'],
'activos_05_02.png'),
bbox_inches='tight')
def analysis_02(start, end, df):
""" """
countries = ['Spain', 'Italy', 'Colombia']
fig, ax = plt.subplots()
fig2, ax2 = plt.subplots()
colors = {'Spain': 'k', 'Colombia': 'r', 'Italy': 'b'}
linewidths = {'Spain': 1, 'Colombia': 2, 'Italy': 1}
start_by_country12 = {
'Spain': ws.dates_keys[10],
'Colombia': ws.dates_keys[44],
'Italy': ws.dates_keys[9]
}
for country in countries:
start = start_by_country12[country]
start_index, end_index = get_start_end(start, end)
data = {}
variables = ['confirmed', 'deaths', 'recovered']
daysmin = 1e99
daysmax = -1e99
for variable in variables:
dates_, x = get_single_time_series(
df[df['Country/Region'] == country], variable, start_index,
end_index)
data[variable] = x[:]
# Existing cases
data['resolved'] = data['recovered'] + data['deaths']
data['existing'] = data['confirmed'] - data['resolved']
days = np.arange(len(dates_))
daysmin = min(days.min(), daysmin)
daysmax = max(days.max(), daysmax)
ax.plot(days,
data['existing'],
'o-',
lw=linewidths[country],
c=colors[country],
alpha=0.5,
label=country)
ax2.plot(days,
data['existing'],
'o-',
lw=linewidths[country],
c=colors[country],
alpha=0.5,
label=country)
utl.configure_axes(ax,
xlims=(daysmin, 20),
xlabel='Days after First Case',
ylabel='New Cases per Day')
utl.configure_axes(ax2,
xlims=(daysmin, 30),
xlabel='Days after First Case',
ylabel='New Cases per Day')
ax.set_ylim(0, 40)
ax.set_title('Existing Cases')
# fig.savefig(os.path.join(ws.folders['website/static/images'], 'activos_01.png'), bbox_inches='tight')
ax2.set_ylim(0, 40)
ax2.set_title('Existing Cases')
def analysis_01(start, end, df):
""" """
countries = ['Spain', 'Italy', 'Colombia']
fig, ax = plt.subplots()
colors = {'Spain': 'k', 'Colombia': 'r', 'Italy': 'b'}
linewidths = {'Spain': 1, 'Colombia': 2, 'Italy': 1}
start_by_country = {
'Spain': ws.dates_keys[10],
'Colombia': ws.dates_keys[44],
'Italy': ws.dates_keys[9]
}
for country in countries:
start = start_by_country[country]
start_index, end_index = get_start_end(start, end)
data = {}
variables = ['confirmed']
daysmin = 1e99
daysmax = -1e99
for variable in variables:
dates_, x = get_single_time_series(
df[df['Country/Region'] == country], variable, start_index,
end_index)
data[variable] = cases_per_day(x)
days = np.arange(len(dates_) - 1)
daysmin = min(days.min(), daysmin)
daysmax = max(days.max(), daysmax)
ax.plot(days,
data[variable],
'o-',
lw=linewidths[country],
c=colors[country],
alpha=0.5,
label=country)
# Date ticks
months = mdates.MonthLocator() # every month
days = mdates.DayLocator() # every month
month_fmt = mdates.DateFormatter('%m')
utl.configure_axes(ax,
xlims=(daysmin, 20),
xlabel='Days after First Case',
ylabel='New Cases per Day')
ax.set_ylim(0, 40)
ax.set_title('Growth Rate (New Cases per Day)')
fig.savefig(os.path.join(ws.folders['website/static/images'],
'ritmos_01.png'),
bbox_inches='tight')
def stackplot(start, end, df):
""" Row-stack graph of the infection """
# Over confirmed
# Absolute
fig, ax = plt.subplots()
start_index, end_index = get_start_end(start, end)
data = {}
variables = ['confirmed', 'recovered', 'deaths']
for variable in variables:
dates_, data[variable] = get_single_time_series(
df, variable, start_index, end_index)
data['resolved'] = data['recovered'] + data['deaths']
data['existing'] = data['confirmed'] - data['resolved']
wanted_variables = ('deaths', 'recovered', 'existing')
values = np.row_stack(np.array([data[k] for k in wanted_variables]))
ax.stackplot(
dates_,
values,
colors=[ws.varcolors[k] for k in wanted_variables],
labels=wanted_variables,
)
ax.plot(dates_,
data['confirmed'],
c=ws.varcolors['confirmed'],
ls='-',
label='confirmed')
utl.configure_axes(ax,
xlims=(dates_[0], dates_[-1]),
xlabel='Date',
ylabel='Cases',
title='Evolution over Confirmed')
fig.savefig(os.path.join(ws.folders['website/static/images'],
'stackplot.png'),
bbox_inches='tight')
# Percentage
fig2, ax2 = plt.subplots()
start_index, end_index = get_start_end(start, end)
data = {}
variables = ['confirmed', 'recovered', 'deaths']
for variable in variables:
dates_, data[variable] = get_single_time_series(
df, variable, start_index, end_index)
data['resolved'] = data['recovered'] + data['deaths']
data['existing'] = data['confirmed'] - data['resolved']
values = np.row_stack(
np.array([data['deaths'], data['recovered'], data['existing']]) *
(100. / data['confirmed']))
ax2.stackplot(
dates_,
values,
colors=[ws.varcolors[k] for k in wanted_variables],
labels=wanted_variables,
)
ax2.set_ylim(0, 100)
utl.configure_axes(ax2,
xlims=(dates_[0], dates_[-1]),
xlabel='Date',
ylabel='Cases (%)',
title='Evolution over Confirmed')
fig2.savefig(os.path.join(ws.folders['website/static/images'],
'stackplot_percentages.png'),
bbox_inches='tight')