updates to notebook from artdgn/pages upstream for phrasing, map, tables, methodology
This commit is contained in:
artdgn
Родитель
d2884e5148
Коммит
ea7ca67da2
Различия файлов скрыты, потому что одна или несколько строк слишком длинны
|
|
@ -14,7 +14,6 @@
|
|||
# name: python3
|
||||
# ---
|
||||
|
||||
# + [markdown] papermill={"duration": 0.016043, "end_time": "2020-04-21T12:16:32.969003", "exception": false, "start_time": "2020-04-21T12:16:32.952960", "status": "completed"} tags=[]
|
||||
# # Projections of ICU need by Country
|
||||
# > Modeling current and future ICU demand.
|
||||
#
|
||||
|
|
@ -25,81 +24,99 @@
|
|||
# - toc: true
|
||||
# - hide: false
|
||||
|
||||
# + [markdown] papermill={"duration": 0.011416, "end_time": "2020-04-21T12:16:32.991324", "exception": false, "start_time": "2020-04-21T12:16:32.979908", "status": "completed"} tags=[]
|
||||
# > Warning: This analysis contains the results of a predictive model that was not prepared or reviewed by an epidemiologist. The assumptions and methods presented should be considered carefully before arriving at any conclusions.
|
||||
# > Warning: This dashboard contains the results of a predictive model that was not built by an epidemiologist.
|
||||
|
||||
# + execution={"iopub.execute_input": "2020-04-21T12:16:33.017500Z", "iopub.status.busy": "2020-04-21T12:16:33.016736Z", "iopub.status.idle": "2020-04-21T12:16:44.018439Z", "shell.execute_reply": "2020-04-21T12:16:44.019026Z"} papermill={"duration": 11.017858, "end_time": "2020-04-21T12:16:44.019212", "exception": false, "start_time": "2020-04-21T12:16:33.001354", "status": "completed"} tags=[]
|
||||
# +
|
||||
#hide
|
||||
import pandas as pd
|
||||
import overview_helpers
|
||||
|
||||
|
||||
helper = overview_helpers.OverviewData
|
||||
stylers = overview_helpers.PandasStyling
|
||||
df_all = helper.table_with_projections()
|
||||
df_all.columns
|
||||
# -
|
||||
|
||||
# + execution={"iopub.execute_input": "2020-04-21T12:16:44.049495Z", "iopub.status.busy": "2020-04-21T12:16:44.048737Z", "iopub.status.idle": "2020-04-21T12:16:44.052619Z", "shell.execute_reply": "2020-04-21T12:16:44.051788Z"} papermill={"duration": 0.021948, "end_time": "2020-04-21T12:16:44.052729", "exception": false, "start_time": "2020-04-21T12:16:44.030781", "status": "completed"} tags=[]
|
||||
#hide_input
|
||||
from IPython.display import Markdown
|
||||
Markdown(f"***Based on data up to: {pd.to_datetime(helper.dt_today).date().isoformat()}***")
|
||||
# -
|
||||
Markdown(f"***Based on data up to: {helper.cur_date}***")
|
||||
|
||||
#hide
|
||||
geo_helper = overview_helpers.GeoMap
|
||||
df_geo = geo_helper.make_geo_df(df_all, cases_filter=1000, deaths_filter=20)
|
||||
fig = geo_helper.make_map_figure(df_geo)
|
||||
|
||||
#hide
|
||||
df_geo['affected_ratio.change.monthly.rate'] = (df_geo['affected_ratio.est.+7d'] -
|
||||
df_geo['affected_ratio.est']) * 30 / 7
|
||||
|
||||
#hide_input
|
||||
fig.update_layout(
|
||||
updatemenus=[
|
||||
dict(
|
||||
buttons=[
|
||||
geo_helper.button_dict(
|
||||
df_geo['needICU.per100k'], 'ICU need<br>(current)',
|
||||
colorscale='Sunsetdark', scale_max=10,
|
||||
subtitle='Estimated current ICU need per 100k population'),
|
||||
df_geo['infection_rate'], 'Transmission rate<br>percent (blue-red)',
|
||||
colorscale='Bluered', scale_max=10, percent=True,
|
||||
subtitle='Transmission rate: over 5% (red) spreading, under 5% (blue) recovering',
|
||||
err_series=df_geo['growth_rate_std']),
|
||||
geo_helper.button_dict(
|
||||
df_geo['needICU.per100k.+14d'], 'ICU need<br>(in 14 days)',
|
||||
colorscale='Sunsetdark', scale_max=10,
|
||||
subtitle='Projected ICU need per 100k population in 14 days',
|
||||
err_series=df_geo['needICU.per100k.+14d.err']),
|
||||
df_geo['infection_rate'], 'Transmission rate<br>percent',
|
||||
colorscale='YlOrRd', scale_max=33, percent=True,
|
||||
subtitle='Transmission rate (related to R0)',
|
||||
err_series=df_geo['growth_rate_std']),
|
||||
geo_helper.button_dict(
|
||||
df_geo['needICU.per100k.+30d'], 'ICU need<br>(in 30 days)',
|
||||
colorscale='Sunsetdark', scale_max=10,
|
||||
subtitle='Projected ICU need per 100k population in 30 days',
|
||||
err_series=df_geo['needICU.per100k.+30d.err']),
|
||||
df_geo['Cases.new.per100k.est'], 'Recent cases<br>estimated per 100k',
|
||||
colorscale='YlOrRd',
|
||||
subtitle='Estimated recent cases in last 5 days per 100k population'),
|
||||
geo_helper.button_dict(
|
||||
df_geo['icu_capacity_per100k'], 'ICU Capacity', colorscale='Blues',
|
||||
subtitle='ICU capacity per 100k population'),
|
||||
df_geo['Cases.new.est'], 'Recent cases<br>(estimated)',
|
||||
colorscale='YlOrRd',
|
||||
subtitle='Estimated recent cases in last 5 days'),
|
||||
geo_helper.button_dict(
|
||||
df_geo['Cases.new.per100k'], 'Recent cases<br>reported per 100k',
|
||||
colorscale='YlOrRd',
|
||||
subtitle='Reported recent cases in last 5 days per 100k population'),
|
||||
geo_helper.button_dict(
|
||||
df_geo['Cases.new'], 'Recent cases<br>(reported)',
|
||||
colorscale='YlOrRd',
|
||||
subtitle='Reported recent cases in last 5 days'),
|
||||
],
|
||||
direction="down", bgcolor='#efdaee',
|
||||
direction="down", bgcolor='#efe9da',
|
||||
pad={"r": 10, "t": 10},
|
||||
showactive=False, x=0.07, xanchor="left", y=1.1, yanchor="top"),
|
||||
dict(
|
||||
buttons=[
|
||||
geo_helper.button_dict(
|
||||
df_geo['affected_ratio.est'], 'Affected percent<br>(Current)',
|
||||
df_geo['affected_ratio.est'], 'Affected percent<br>(Current)',
|
||||
colorscale='Bluyl', percent=True,
|
||||
subtitle='Estimated current affected population percentage'),
|
||||
geo_helper.button_dict(
|
||||
df_geo['affected_ratio.est.+14d'], 'Affected percent<br>(in 14 days)',
|
||||
df_geo['affected_ratio.est.+14d'], 'Affected percent<br>(in 14 days)',
|
||||
colorscale='Bluyl', scale_max=25, percent=True,
|
||||
subtitle='Projected affected population percentage in 14 days',
|
||||
err_series=df_geo['affected_ratio.est.+14d.err']),
|
||||
geo_helper.button_dict(
|
||||
df_geo['affected_ratio.est.+30d'], 'Affected percent<br>(in 30 days)',
|
||||
df_geo['affected_ratio.est.+30d'], 'Affected percent<br>(in 30 days)',
|
||||
colorscale='Bluyl', scale_max=25, percent=True,
|
||||
subtitle='Projected affected population percentage in 30 days',
|
||||
err_series=df_geo['affected_ratio.est.+30d.err']),
|
||||
geo_helper.button_dict(
|
||||
df_geo['Cases.total.per100k.est'], 'Total cases<br>estimated per 100k',
|
||||
df_geo['affected_ratio.change.monthly.rate'],
|
||||
title='Affected percent<br>montly change rate',
|
||||
colorscale='Bluyl', scale_max=10, percent=True,
|
||||
subtitle='Current affected population percentage monthly change rate',
|
||||
err_series=df_geo['affected_ratio.est.+30d.err']),
|
||||
geo_helper.button_dict(
|
||||
df_geo['Cases.total.per100k.est'], 'Total cases<br>estimated per 100k',
|
||||
colorscale='YlOrRd',
|
||||
subtitle='Estimated total cases per 100k population'),
|
||||
geo_helper.button_dict(
|
||||
df_geo['Cases.total.est'], 'Total cases<br>(estimated)', colorscale='YlOrRd',
|
||||
subtitle='Estimated total cases'),
|
||||
geo_helper.button_dict(
|
||||
df_geo['Cases.total.per100k'], 'Total cases<br>reported per 100k',
|
||||
df_geo['Cases.total.per100k'], 'Total cases<br>reported per 100k',
|
||||
colorscale='YlOrRd',
|
||||
subtitle='Reported total cases per 100k population'),
|
||||
geo_helper.button_dict(
|
||||
|
|
@ -108,39 +125,31 @@ fig.update_layout(
|
|||
],
|
||||
direction="down", bgcolor='#dceae1',
|
||||
pad={"r": 10, "t": 10},
|
||||
showactive=False, x=0.24, xanchor="left", y=1.1, yanchor="top"),
|
||||
showactive=False, x=0.29, xanchor="left", y=1.1, yanchor="top"),
|
||||
dict(
|
||||
buttons=[
|
||||
geo_helper.button_dict(
|
||||
df_geo['infection_rate'], 'Infection rate<br>percent (blue-red)',
|
||||
colorscale='Bluered', scale_max=10, percent=True,
|
||||
subtitle='Infection spread rate: over 5% (red) spreading, under 5% (blue) recovering',
|
||||
err_series=df_geo['growth_rate_std']),
|
||||
df_geo['needICU.per100k'], 'ICU need<br>(current)',
|
||||
colorscale='Sunsetdark', scale_max=10,
|
||||
subtitle='Estimated current ICU need per 100k population'),
|
||||
geo_helper.button_dict(
|
||||
df_geo['infection_rate'], 'Infection rate<br>percent',
|
||||
colorscale='YlOrRd', scale_max=33, percent=True,
|
||||
subtitle='Infection spread rate (related to R0)',
|
||||
err_series=df_geo['growth_rate_std']),
|
||||
df_geo['needICU.per100k.+14d'], 'ICU need<br>(in 14 days)',
|
||||
colorscale='Sunsetdark', scale_max=10,
|
||||
subtitle='Projected ICU need per 100k population in 14 days',
|
||||
err_series=df_geo['needICU.per100k.+14d.err']),
|
||||
geo_helper.button_dict(
|
||||
df_geo['Cases.new.per100k.est'], 'New cases<br>estimated per 100k',
|
||||
colorscale='YlOrRd',
|
||||
subtitle='Estimated new cases in last 5 days per 100k population'),
|
||||
df_geo['needICU.per100k.+30d'], 'ICU need<br>(in 30 days)',
|
||||
colorscale='Sunsetdark', scale_max=10,
|
||||
subtitle='Projected ICU need per 100k population in 30 days',
|
||||
err_series=df_geo['needICU.per100k.+30d.err']),
|
||||
geo_helper.button_dict(
|
||||
df_geo['Cases.new.est'], 'New cases<br>(estimated)',
|
||||
colorscale='YlOrRd',
|
||||
subtitle='Estimated new cases in last 5 days'),
|
||||
geo_helper.button_dict(
|
||||
df_geo['Cases.new.per100k'], 'New cases<br>reported per 100k',
|
||||
colorscale='YlOrRd',
|
||||
subtitle='Reported new cases in last 5 days per 100k population'),
|
||||
geo_helper.button_dict(
|
||||
df_geo['Cases.new'], 'New cases<br>(reported)',
|
||||
colorscale='YlOrRd',
|
||||
subtitle='Reported new cases in last 5 days'),
|
||||
df_geo['icu_capacity_per100k'], 'Pre-COVID<br>ICU Capacity',
|
||||
colorscale='Blues',
|
||||
subtitle='Pre-COVID ICU capacity per 100k population'),
|
||||
],
|
||||
direction="down", bgcolor='#efe9da',
|
||||
direction="down", bgcolor='#efdaee',
|
||||
pad={"r": 10, "t": 10},
|
||||
showactive=False, x=0.46, xanchor="left", y=1.1, yanchor="top"),
|
||||
showactive=False, x=0.515, xanchor="left", y=1.1, yanchor="top"),
|
||||
dict(
|
||||
buttons=[
|
||||
geo_helper.button_dict(
|
||||
|
|
@ -150,42 +159,39 @@ fig.update_layout(
|
|||
df_geo['Deaths.total'], 'Deaths<br>Total', colorscale='Reds',
|
||||
subtitle='Total deaths'),
|
||||
geo_helper.button_dict(
|
||||
df_geo['Deaths.new.per100k'], 'New deaths<br>per 100k', colorscale='Reds',
|
||||
subtitle='New deaths in last 5 days per 100k population'),
|
||||
df_geo['Deaths.new.per100k'], 'Recent deaths<br>per 100k', colorscale='Reds',
|
||||
subtitle='Recent deaths in last 5 days per 100k population'),
|
||||
geo_helper.button_dict(
|
||||
df_geo['Deaths.new'], 'New deaths<br>total', colorscale='Reds',
|
||||
subtitle='New deaths in last 5 days'),
|
||||
df_geo['Deaths.new'], 'Recent deaths<br>total', colorscale='Reds',
|
||||
subtitle='Recent deaths in last 5 days'),
|
||||
geo_helper.button_dict(
|
||||
df_geo['lagged_fatality_rate'], 'Fatality rate %<br>(lagged)',
|
||||
df_geo['lagged_fatality_rate'], 'Fatality rate %<br>(lagged)',
|
||||
colorscale='Reds', scale_max=20, percent=True,
|
||||
subtitle='Reported fatality rate (relative to reported cases 8 days ago)'),
|
||||
],
|
||||
direction="down", bgcolor='#efdbda',
|
||||
pad={"r": 10, "t": 10},
|
||||
showactive=False, x=0.68, xanchor="left", y=1.1, yanchor="top"),
|
||||
showactive=False, x=0.69, xanchor="left", y=1.1, yanchor="top"),
|
||||
]);
|
||||
|
||||
# # World map (interactive)
|
||||
# > Includes only countries with at least 1000 reported cases or at least 20 reported deaths.
|
||||
#
|
||||
# - Details of estimation and prediction calculations are in [Appendix](#methodology) and in [Tables](#tables), as well as [Plots of model predictions](#examples).
|
||||
# - Details of estimation and prediction calculations are in [Appendix](#appendix) and in [Tables](#tables), as well as [Plots of model predictions](#examples).
|
||||
# - New cases and new deaths refer to cases or deaths in the last 5 days.
|
||||
|
||||
# > Tip: Select columns to show on map to from the dropdown menus. The map is zoomable and draggable.
|
||||
|
||||
#hide_input
|
||||
# from IPython.display import HTML
|
||||
# HTML(fig.to_html())
|
||||
fig.show()
|
||||
|
||||
# + [markdown] papermill={"duration": 0.01049, "end_time": "2020-04-21T12:16:44.072280", "exception": false, "start_time": "2020-04-21T12:16:44.061790", "status": "completed"} tags=[]
|
||||
# # Tables
|
||||
# <a id='tables'></a>
|
||||
# ## Projected need for ICU beds
|
||||
#
|
||||
# > Countries sorted by current ICU demand, split into Growing and Recovering countries by current infection rate.
|
||||
# > Countries sorted by current ICU demand, split into Growing and Recovering countries by current transmission rate.
|
||||
#
|
||||
# - Details of estimation and prediction calculations are in [Appendix](#methodology), as well as [Plots of model predictions](#examples).
|
||||
# - Details of estimation and prediction calculations are in [Appendix](#appendix), as well as [Plots of model predictions](#examples).
|
||||
# - Column definitions:
|
||||
# - <font size=2><b>Estimated ICU need per 100k population</b>: number of ICU beds estimated to be needed per 100k population by COVID-19 patents.</font>
|
||||
# - <font size=2><b>Estimated daily infection rate</b>: daily percentage rate of new infections relative to active infections during last 5 days.</font>
|
||||
|
|
@ -194,14 +200,14 @@ fig.show()
|
|||
# - <font size=2><b>ICU capacity per 100k</b>: number of ICU beds per 100k population.</font>
|
||||
# - <font size=2><b>Estimated ICU Spare capacity per 100k</b>: estimated ICU capacity per 100k population based on assumed normal occupancy rate of 70% and number of ICU beds (only for countries with ICU beds data).</font>
|
||||
|
||||
# + [markdown] papermill={"duration": 0.009167, "end_time": "2020-04-21T12:16:44.092197", "exception": false, "start_time": "2020-04-21T12:16:44.083030", "status": "completed"} tags=[]
|
||||
# > Tip: The <b><font color="b21e3e">red (need for ICU)</font></b> and the <b><font color="3ab1d8">blue (ICU spare capacity)</font></b> bars are on the same 0-10 scale, for easy visual comparison of columns.
|
||||
# -
|
||||
|
||||
#hide
|
||||
df = helper.filter_df(df_all)
|
||||
df_filt = helper.filter_df(df_all)
|
||||
df = df_filt.rename(index={'Bosnia and Herzegovina': 'Bosnia',
|
||||
'United Arab Emirates': 'UAE'})
|
||||
|
||||
# + execution={"iopub.execute_input": "2020-04-21T12:16:44.124650Z", "iopub.status.busy": "2020-04-21T12:16:44.119771Z", "iopub.status.idle": "2020-04-21T12:16:44.991670Z", "shell.execute_reply": "2020-04-21T12:16:44.992136Z"} papermill={"duration": 0.890669, "end_time": "2020-04-21T12:16:44.992275", "exception": false, "start_time": "2020-04-21T12:16:44.101606", "status": "completed"} tags=[]
|
||||
# +
|
||||
#hide
|
||||
df_data = df.sort_values('needICU.per100k', ascending=False)
|
||||
df_pretty = df_data.copy()
|
||||
|
|
@ -212,11 +218,11 @@ df_pretty['needICU.per100k.+30d'] = stylers.with_errs_float(
|
|||
df_pretty['infection_rate'] = stylers.with_errs_ratio(df_pretty, 'infection_rate', 'growth_rate_std')
|
||||
|
||||
cols = {'needICU.per100k': 'Estimated<br>current<br>ICU need<br>per 100k<br>population',
|
||||
'infection_rate': 'Estimated<br>daily infection<br>rate',
|
||||
'needICU.per100k.+14d': 'Projected<br>ICU need<br>per 100k<br>In 14 days',
|
||||
'needICU.per100k.+30d': 'Projected<br>ICU need<br>per 100k<br>In 30 days',
|
||||
'icu_capacity_per100k': 'ICU<br>capacity<br> per 100k',
|
||||
'icu_spare_capacity_per100k': 'Estimated ICU<br>Spare capacity<br>per 100k',
|
||||
'infection_rate': 'Estimated<br>daily<br>transmission<br>rate',
|
||||
'needICU.per100k.+14d': 'Projected<br>ICU need<br>per 100k<br>In 14 days',
|
||||
'needICU.per100k.+30d': 'Projected<br>ICU need<br>per 100k<br>In 30 days',
|
||||
'icu_capacity_per100k': 'Pre-COVID<br>ICU<br>capacity<br> per 100k',
|
||||
'icu_spare_capacity_per100k': 'Pre-COVID<br>Estimated ICU<br>Spare capacity<br>per 100k',
|
||||
}
|
||||
|
||||
def style_icu_table(df_pretty, filt):
|
||||
|
|
@ -237,48 +243,44 @@ def style_icu_table(df_pretty, filt):
|
|||
|
||||
# -
|
||||
|
||||
# ### Growing countries (infection rate above 5%)
|
||||
# ### Growing countries (transmission rate above 5%)
|
||||
|
||||
#hide_input
|
||||
style_icu_table(df_pretty, df_data['infection_rate'] > 0.05)
|
||||
|
||||
# ### Recovering countries (infection rate below 5%)
|
||||
# ### Recovering countries (tranmission rate below 5%)
|
||||
|
||||
#hide_input
|
||||
style_icu_table(df_pretty, df_data['infection_rate'] <= 0.05)
|
||||
|
||||
# + [markdown] papermill={"duration": 0.013253, "end_time": "2020-04-21T12:16:45.018802", "exception": false, "start_time": "2020-04-21T12:16:45.005549", "status": "completed"} tags=[]
|
||||
# # Appendix
|
||||
# <a id='appendix'></a>
|
||||
|
||||
# + [markdown] papermill={"duration": 0.013516, "end_time": "2020-04-21T12:16:45.047181", "exception": false, "start_time": "2020-04-21T12:16:45.033665", "status": "completed"} tags=[]
|
||||
# <a id='examples'></a>
|
||||
#
|
||||
# ## Interactive plot of Model predictions
|
||||
# ## Interactive plot of model predictions and past data
|
||||
#
|
||||
# > Tip: Choose a country from the drop-down menu to see the calculations used in the tables above and the dynamics of the model.
|
||||
|
||||
# + execution={"iopub.execute_input": "2020-04-21T12:16:45.087905Z", "iopub.status.busy": "2020-04-21T12:16:45.087062Z", "iopub.status.idle": "2020-04-21T12:17:00.351387Z", "shell.execute_reply": "2020-04-21T12:17:00.351941Z"} papermill={"duration": 15.291152, "end_time": "2020-04-21T12:17:00.352091", "exception": false, "start_time": "2020-04-21T12:16:45.060939", "status": "completed"} tags=[]
|
||||
#hide_input
|
||||
_, debug_dfs = helper.table_with_projections(debug_dfs=True)
|
||||
df_alt = pd.concat([d.reset_index() for d in debug_dfs], axis=0)
|
||||
overview_helpers.altair_sir_plot(df_alt, df['needICU.per100k.+14d.min'].idxmax())
|
||||
df_alt_filt = df_alt[(df_alt['day'] > -60) & (df_alt['country'].isin(df_filt.index))]
|
||||
overview_helpers.altair_sir_plot(df_alt_filt, df_filt['Deaths.new.per100k'].idxmax())
|
||||
|
||||
# + [markdown] papermill={"duration": 0.120423, "end_time": "2020-04-21T12:17:00.635637", "exception": false, "start_time": "2020-04-21T12:17:00.515214", "status": "completed"} tags=[]
|
||||
# ## Projected Affected Population percentage
|
||||
# > Countries sorted by number of new cases in last 5 days. The projected affected population percentage is directly related to the calculation of estimated ICU need.
|
||||
|
||||
# + [markdown] papermill={"duration": 0.127632, "end_time": "2020-04-21T12:17:00.889117", "exception": false, "start_time": "2020-04-21T12:17:00.761485", "status": "completed"} tags=[]
|
||||
# - Details of estimation and prediction calculations are in [Appendix](#methodology), as well as [Plots of model predictions](#examples).
|
||||
# ## Projected Affected Population percentages
|
||||
# > Top 20 countries with most estimated recent cases.
|
||||
#
|
||||
# - Sorted by number of estimated recent cases during the last 5 days.
|
||||
# - Column definitions:
|
||||
# - <font size=2><b>Estimated <i>new</i> cases in last 5 days</b>: self explanatory.</font>
|
||||
# - <font size=2><b>Estimated <i>total</i> affected population percentage</b>: estimated percentage of total population already affected (infected, recovered, or dead).</font>
|
||||
# - <font size=2><b>Estimated daily infection rate</b>: daily percentage rate of new infections relative to active infections during last 5 days.</font>
|
||||
# - <font size=2><b>Projected total affected percentage in 14 days</b>: of population.</font>
|
||||
# - <font size=2><b>Projected total affected percentage in 30 days</b>: of population.</font>
|
||||
# - <font size=2><b>Lagged fatality rate</b>: reported total deaths divided by total cases 8 days ago.</font>
|
||||
# - <font size=2><b>Estimated <i>recent</i> cases in last 5 days</b>: self explanatory.</font>
|
||||
# - <font size=2><b>Estimated <i>total</i> affected population percentage</b>: estimated percentage of total population already affected (infected, recovered, or dead).</font>
|
||||
# - <font size=2><b>Estimated daily tranmission rate</b>: daily percentage rate of recent infections relative to active infections during last 5 days.</font>
|
||||
# - <font size=2><b>Projected total affected percentage in 14 days</b>: of population.</font>
|
||||
# - <font size=2><b>Projected total affected percentage in 30 days</b>: of population.</font>
|
||||
# - <font size=2><b>Lagged fatality rate</b>: reported total deaths divided by total cases 8 days ago.</font>
|
||||
|
||||
# + execution={"iopub.execute_input": "2020-04-21T12:17:01.170022Z", "iopub.status.busy": "2020-04-21T12:17:01.169348Z", "iopub.status.idle": "2020-04-21T12:17:01.631320Z", "shell.execute_reply": "2020-04-21T12:17:01.631845Z"} papermill={"duration": 0.60958, "end_time": "2020-04-21T12:17:01.632016", "exception": false, "start_time": "2020-04-21T12:17:01.022436", "status": "completed"} tags=[]
|
||||
# +
|
||||
#hide_input
|
||||
df_data = df.sort_values('Cases.new.est', ascending=False).head(20)
|
||||
df_pretty = df_data.copy()
|
||||
|
|
@ -290,10 +292,10 @@ df_pretty['infection_rate'] = stylers.with_errs_ratio(df_pretty, 'infection_rate
|
|||
|
||||
cols = {'Cases.new.est': 'Estimated <br> <i>new</i> cases <br> in last 5 days',
|
||||
'affected_ratio.est': 'Estimated <br><i>total</i><br>affected<br>population<br>percentage',
|
||||
'infection_rate': 'Estimated <br>daily infection<br>rate',
|
||||
'infection_rate': 'Estimated<br>daily<br>tranmission<br>rate',
|
||||
'affected_ratio.est.+14d': 'Projected<br><i>total</i><br>affected<br>percentage<br>In 14 days',
|
||||
'affected_ratio.est.+30d': 'Projected<br><i>total</i><br>affected<br>percentage<br>In 30 days',
|
||||
'lagged_fatality_rate': 'Lagged<br>fatality<br>rate',
|
||||
'affected_ratio.est.+30d': 'Projected<br><i>total</i><br>affected<br>percentage<br>In 30 days',
|
||||
'lagged_fatality_rate': 'Lagged<br>fatality <br> percentage',
|
||||
}
|
||||
|
||||
df_pretty[cols.keys()].rename(cols, axis=1).style\
|
||||
|
|
@ -309,24 +311,25 @@ df_pretty[cols.keys()].rename(cols, axis=1).style\
|
|||
.applymap(lambda _: 'color: red', subset=cols['lagged_fatality_rate'])\
|
||||
.format('<b>{:,.0f}</b>', subset=cols['Cases.new.est'])\
|
||||
.format('<b>{:.1%}</b>', subset=[cols['lagged_fatality_rate'], cols['affected_ratio.est']])
|
||||
# -
|
||||
|
||||
# + [markdown] papermill={"duration": 0.12762, "end_time": "2020-04-21T12:17:01.880996", "exception": false, "start_time": "2020-04-21T12:17:01.753376", "status": "completed"} tags=[]
|
||||
# ## Methodology & Assumptions
|
||||
# <a id='methodology'></a>
|
||||
# ## Methodology
|
||||
# - I'm not an epidemiologist. This is an attempt to understand what's happening, and what the future looks like if current trends remain unchanged.
|
||||
# - Everything is approximated and depends heavily on underlying assumptions.
|
||||
# - Countries with less than 10 total deaths or less than 1 Million population are excluded.
|
||||
# - Projection is done using a simple [SIR model](https://en.wikipedia.org/wiki/Compartmental_models_in_epidemiology#The_SIR_model) with (see [examples](#examples)) combined with the approach in [Total Outstanding Cases](https://covid19dashboards.com/outstanding_cases/#Appendix:-Methodology-of-Predicting-Recovered-Cases):
|
||||
# - Growth rate calculated over the 5 past days. This is pessimistic - because it includes the testing rate growth rate as well, and is slow to react to both improvements in test coverage and "flattening" due to social isolation.
|
||||
# - Confidence bounds are calculated by from the weighted STD of the growth rate over the last 5 days. Model predictions are calculated for growth rates within 1 STD of the weighted mean. The maximum and minimum values for each day are used as confidence bands.
|
||||
# - For projections (into future) very noisy projections (with broad confidence bounds) are not shown in the tables.
|
||||
# - Recovery probability being 1/20 (for 20 days to recover) where the rate estimated from [Total Outstanding Cases](https://covid19dashboards.com/outstanding_cases/#Appendix:-Methodology-of-Predicting-Recovered-Cases) is too high (on down-slopes).
|
||||
# - ICU need is calculated as being [4.4% of active reported cases](https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College-COVID19-NPI-modelling-16-03-2020.pdf) where:
|
||||
# - Active cases are taken from the SIR model. The ICU need is calculated from reported cases rather than from total estimated active cases. This is because the ICU ratio (4.4%) is based on symptomatic reported cases.
|
||||
# - ICU capacities are from [Wikipedia](https://en.wikipedia.org/wiki/List_of_countries_by_hospital_beds) (OECD countries mostly) and [CCB capacities in Asia](https://www.researchgate.net/publication/338520008_Critical_Care_Bed_Capacity_in_Asian_Countries_and_Regions).
|
||||
# - ICU spare capacity is based on 70% normal occupancy rate ([66% in US](https://www.sccm.org/Blog/March-2020/United-States-Resource-Availability-for-COVID-19), [75% OECD](https://www.oecd-ilibrary.org/social-issues-migration-health/health-at-a-glance-2019_4dd50c09-en))
|
||||
# - Total case estimation calculated from deaths by:
|
||||
# - Assuming that unbiased fatality rate is 0.72% ([current meta estimate in](https://www.cebm.net/covid-19/global-covid-19-case-fatality-rates/)).
|
||||
# - Recovery probability being 1/20 (for 20 days to recover) where the rate estimated from [Total Outstanding Cases](https://covid19dashboards.com/outstanding_cases/#Appendix:-Methodology-of-Predicting-Recovered-Cases) is too high (on down-slopes).
|
||||
# - Total cases are estimated from deaths in each country:
|
||||
# - Each country has different testing policy and capacity and cases are under-reported in some countries. Using an estimated IFR (fatality rate) we can estimate the number of cases some time ago by using the total deaths until today. We can than use this estimation to estimate the testing bias and multiply the current numbers by that.
|
||||
# - IFRs for each country is estimated using the age IFRs from [May 1 New York paper](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3590771) and [UN demographic data for 2020](https://population.un.org/wpp/Download/Standard/Population/). These IFRs can be found in `df['age_adjusted_ifr']` column. Some examples: US - 0.98%, UK - 1.1%, Qatar - 0.25%, Italy - 1.4%, Japan - 1.6%.
|
||||
# - The average fatality lag is assumed to be 8 days on average for a case to go from being confirmed positive (after incubation + testing lag) to death. This is the same figure used by ["Estimating The Infected Population From Deaths"](https://covid19dashboards.com/covid-infected/).
|
||||
# - Testing bias: the actual lagged fatality rate is than divided by the 0.72% figure to estimate the testing bias in a country. The estimated testing bias then multiplies the reported case numbers to estimate the *true* case numbers (*=case numbers if testing coverage was as comprehensive as in the heavily tested countries*).
|
||||
# - The testing bias calculation is a high source of uncertainty in all these estimations and projections. Better source of testing bias (or just *true case* numbers), should make everything more accurate.
|
||||
# - Testing bias: the actual lagged fatality rate is than divided by the IFR to estimate the testing bias in a country. The estimated testing bias then multiplies the reported case numbers to estimate the *true* case numbers (*=case numbers if testing coverage was as comprehensive as in the heavily tested countries*).
|
||||
# - ICU need is calculated and age-adjusted as follows:
|
||||
# - UK ICU ratio was reported as [4.4% of active reported cases](https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College-COVID19-NPI-modelling-16-03-2020.pdf).
|
||||
# - Using UKs ICU ratio and IFRs corrected for age demographics we can estimate each country's ICU ratio (the number of cases requiring ICU hospitalisation). For example using the IFR ratio between UK and Qatar to devide UK's 4.4% we get an ICU ratio of around 1% for Qatar which is also the ratio [they report to WHO here](https://apps.who.int/gb/COVID-19/pdf_files/30_04/Qatar.pdf).
|
||||
# - Active cases are taken from the SIR model. The ICU need is calculated from reported cases rather than from total estimated active cases. This is because the ICU ratio (4.4%) is based on reported cases.
|
||||
# - Pre COVID-19 ICU capacities are from [Wikipedia](https://en.wikipedia.org/wiki/List_of_countries_by_hospital_beds) (OECD countries mostly) and [CCB capacities in Asia](https://www.researchgate.net/publication/338520008_Critical_Care_Bed_Capacity_in_Asian_Countries_and_Regions).
|
||||
# - Pre COVID-19 ICU spare capacity is based on 70% normal occupancy rate ([66% in US](https://www.sccm.org/Blog/March-2020/United-States-Resource-Availability-for-COVID-19), [75% OECD](https://www.oecd-ilibrary.org/social-issues-migration-health/health-at-a-glance-2019_4dd50c09-en))
|
||||
#
|
||||
|
|
|
|||
|
|
@ -14,6 +14,7 @@ pd.set_option('display.max_rows', None)
|
|||
pd.set_option('display.max_columns', None)
|
||||
pd.set_option('display.width', 10000)
|
||||
|
||||
SAVE_JHU_DATA = False
|
||||
|
||||
class SourceData:
|
||||
df_mappings = pd.read_csv(os.path.join(data_folder, 'mapping_countries.csv'))
|
||||
|
|
@ -47,7 +48,8 @@ class SourceData:
|
|||
@classmethod
|
||||
def get_covid_dataframe(cls, name):
|
||||
df = cls._download_covid_df(name)
|
||||
cls._save_covid_df(df, name)
|
||||
if SAVE_JHU_DATA:
|
||||
cls._save_covid_df(df, name)
|
||||
|
||||
# rename countries
|
||||
df[COL_REGION] = df[COL_REGION].replace(cls.mappings['replace.country'])
|
||||
|
|
@ -691,8 +693,8 @@ class GeoMap:
|
|||
def make_map_figure(cls,
|
||||
df_plot_geo,
|
||||
col='infection_rate',
|
||||
title='Transmition rate<br>percent (blue-red)',
|
||||
subtitle='Transmition rate: over 5% (red) '
|
||||
title='Transmission rate<br>percent (blue-red)',
|
||||
subtitle='Transmission rate: over 5% (red) '
|
||||
'spreading, under 5% (blue) recovering'):
|
||||
import plotly.graph_objects as go
|
||||
|
||||
|
|
@ -702,7 +704,7 @@ class GeoMap:
|
|||
f"Cases (reported): {r['Cases.total']:,.0f} (+<b>{r['Cases.new']:,.0f}</b>)<br>"
|
||||
f"Cases (estimated): {r['Cases.total.est']:,.0f} (+<b>{r['Cases.new.est']:,.0f}</b>)<br>"
|
||||
f"Affected percent: <b>{r['affected_ratio.est']:.1%}</b><br>"
|
||||
f"Transmition rate: <b>{r['infection_rate']:.1%}</b> ± {r['growth_rate_std']:.1%}<br>"
|
||||
f"Transmission rate: <b>{r['infection_rate']:.1%}</b> ± {r['growth_rate_std']:.1%}<br>"
|
||||
f"Deaths: {r['Deaths.total']:,.0f} (+<b>{r['Deaths.new']:,.0f}</b>)<br>"
|
||||
), axis=1))
|
||||
|
||||
|
|
|
|||
|
|
@ -33,6 +33,7 @@ services:
|
|||
- "4000:4000"
|
||||
volumes:
|
||||
- .:/data/
|
||||
dns: 8.8.8.8
|
||||
command: >
|
||||
bash -c "gem install bundler
|
||||
&& jekyll serve --trace"
|
||||
|
|
|
|||
Загрузка…
Ссылка в новой задаче