US Census

Add US census data to the atlas

notebooks/examples/italy-covid-19.ipynb

 %% Cell type:markdown id: tags:
 # COVID-19 Case data for Italy
 Data from [Civil Protection of Italy](https://github.com/pcm-dpc/COVID-19).
 %% Cell type:code id: tags:
 ``` python
 %load_ext autoreload
 %autoreload 2
 ```
 %% Cell type:code id: tags:
 ``` python
-import ipywidgets as widgets
-
 import pandas as pd
 import altair as alt
 from IPython.display import display, HTML
 from covid_19_dashboard import helper, plotting
 from covid_19_dashboard.italy_utils import get_region_populations, prepare_dataframe
 ```
 %% Cell type:code id: tags:
 ``` python
 html_credits=HTML('''
 <p style="font-size: smaller">Data Sources:
   <a href="https://github.com/pcm-dpc/COVID-19">Italian Civil Protection</a>, <a href="https://wikidata.org">Wikidata</a>
 <br>
 Analysis and Visualization:
   <a href="https://renkulab.io/projects/covid-19/covid-19-public-data">Covid-19 Public Data Collaboration Project</a>
 </p>''')
 ```
 %% Cell type:code id: tags:parameters
 ``` python
 data_folder = "../../data/covid-19-italy/"
 ```
 %% Cell type:code id: tags:
 ``` python
 df_national = prepare_dataframe(
     data_folder,
     "dpc-covid19-ita-andamento-nazionale.csv",
     "dati-andamento-nazionale-description.json",
     use_time_index=True
 )
 ```
 %% Cell type:code id: tags:
 ``` python
 # compute daily new cases
 df_national.reset_index(level=0, inplace=True)
 df_national["New cases per day"] = df_national["total_cases"].diff()
 ```
 %% Cell type:markdown id: tags:
 ## National summary of total cases and tests
 %% Cell type:code id: tags:
 ``` python
 # make the charts
 base = alt.Chart(df_national).mark_bar(size=10).encode(alt.X('date'))
 daily = base.encode(alt.Y('New cases per day'))
 total = base.mark_point(color='orange', filled=True, size=100, opacity=1).encode(alt.Y('total_cases', title="Cases (orange)"))
 tests = base.encode(alt.Y('tests', title="Tests (blue)"))
 totals = alt.layer(tests, total)
 chart = alt.hconcat(totals, daily,
             title='Total and daily tests and cases nationwide'
 ).configure_title(
     anchor='middle'
 )
 display(chart)
 display(html_credits)
 ```
 %% Output
 %% Cell type:code id: tags:
 ``` python
 # set up the regional data frame
 df_regions = prepare_dataframe(
     data_folder,
     "dpc-covid19-ita-regioni.csv",
     "dati-regioni-description.json",
     use_time_index=True
 )
 df_regions.reset_index(level=0, inplace=True)
 # The regional data includes Bolzano and Trento as separate regions - they are two provinces of the Trentino Alto Adige region, so merge them together
 df_trentino_alto_adige = df_regions.loc[df_regions['region_code'] == 4].groupby('date').sum().reset_index()
 df_trentino_alto_adige['region_code'] = 4
 df_trentino_alto_adige['latitude'] = 46.4337
 df_trentino_alto_adige['longitude'] = 11.1693
 df_trentino_alto_adige['region'] = 'Trentino Alto Adige'
 df_trentino_alto_adige['country_code'] = 'ITA'
 df_regions = df_regions[df_regions['region_code'] !=4 ].append(df_trentino_alto_adige)
 ```
 %% Cell type:code id: tags:
 ``` python
 # Retrieve population data and include it in the dataframe
 populations = get_region_populations()
 for field in ['total_cases', 'tests', 'dead']:
     df_regions[f'{field}_100k'] = df_regions[[field,'region_code']].apply(lambda row: row[field]/populations[row['region_code']]*100000, axis=1)
 ```
 %% Cell type:code id: tags:
 ``` python
 max_cases = df_regions.set_index('region').max(level='region')['total_cases'].reset_index()
 sorted_regions = max_cases.sort_values(by='total_cases', ascending=False)['region']
 ```
 %% Cell type:markdown id: tags:
 ## Cases per region
 %% Cell type:code id: tags:
 ``` python
 # Total cases and deaths per 100k population
 base = alt.Chart(df_regions).encode(alt.X('date'))
 death = base.mark_line(color='black').encode(alt.Y('dead', title='Deaths'))
 total = base.mark_bar(color='orange', size=5).encode(alt.Y('total_cases', title="Total cases")).properties(
     width=150,
     height=150)
 totals = alt.layer(total, death)
 chart = totals.facet(facet='region',
              columns=5,
              align='each',
              title='Total cases (orange) and deaths (black) by region'
 ).configure_title(
     anchor='middle'
 )
 display(chart)
 display(html_credits)
 ```
 %% Output
 %% Cell type:code id: tags:
 ``` python
 # Total cases and deaths per 100k population
 base = alt.Chart(df_regions).encode(alt.X('date'))
 death = base.mark_line(color='black').encode(alt.Y('dead_100k', title='Deaths'))
 total = base.mark_bar(color='orange', size=5).encode(alt.Y('total_cases_100k', title="Total cases")).properties(
     width=150,
     height=150)
 totals = alt.layer(total, death)
 chart = totals.facet(facet='region',
              columns=5,
              align='each',
              title='Total cases (orange) and deaths (black) by region per 100k population'
 ).configure_title(
     anchor='middle'
 )
 display(chart)
 display(html_credits)
 ```
 %% Output
 %% Cell type:code id: tags:
 ``` python
 # compute daily differences
 tdf = df_regions.sort_values(['region', 'date'], ascending=[True, False]).set_index(['region', 'date'])
 diffs_df = tdf[['total_cases', 'dead']].groupby(level='region').diff(periods=-1).dropna(how='all')
 tdf_diff=tdf.join(diffs_df, rsuffix='_diff').reset_index()
 ```
 %% Cell type:code id: tags:
 ``` python
 # make the daily charts
 base = alt.Chart(tdf_diff).encode(alt.X('date')).properties(
     width=150,
     height=150,
 )
 daily_new = base.mark_bar(color='orange').encode(alt.Y('total_cases_diff', title='N'))
 daily_dead = base.mark_line(color='black').encode(alt.Y('dead_diff'))
 chart = (daily_new + daily_dead).facet(
     facet='region',
     columns=5,
     align='each',
     title='Daily new cases (orange) and deaths (black) by region',
 ).configure_title(
     anchor='middle'
 )
 display(chart)
 display(html_credits)
 ```
 %% Output
 %% Cell type:code id: tags:
 ``` python
 ```
 %% Cell type:markdown id: tags:
 ## Rates of growth
 How quickly is the infection spreading? The charts below show data for all of the regions of Italy framed by some illustrative guiding lines to give an idea of how the rate of infection is evolving.
 %% Cell type:code id: tags:
 ``` python
 since_100th_case = helper.make_since_df(df_regions, column='total_cases', region_column='region')
 base = alt.Chart(since_100th_case, title="Italy: total cases since 100th case").mark_line(interpolate='basis')
 line_chart = plotting.make_region_since_chart(
     base,
     'total_cases',
     'sinceDay0',
     'region',
     'Days since 100th case',
     'Cumulative cases',
     'Cases',
     'Region'
 ).properties(
     width=450,
     height=450
 )
 rule_chart = plotting.make_rule_chart(max_case=110000, max_days=40, pos_day=(12,50000), pos_3days=(35,300000), pos_week=(40,2000))
 # add the total cases for italy to the chart
 since_100th_case_all = helper.make_since_df(df_national, start_case=100, column='total_cases', region_column='country_code')
 since_100th_case_all['country_label'] = 'Italy'
 italy = alt.Chart(since_100th_case_all).mark_line(color='steelblue').encode(x='sinceDay0', y='total_cases')
 italy += italy.mark_text(dy=-10).encode(text='country_label').transform_filter(f"datum['sinceDay0'] == {len(since_100th_case_all)-1}")
 line_chart + rule_chart + italy
 ```
 %% Output
     alt.LayerChart(...)
 %% Cell type:code id: tags:
 ``` python
 since_10th_death = helper.make_since_df(df_regions, start_case=10, column='dead', region_column='region')
 base = alt.Chart(since_10th_death, title="Italy: total deaths since 10th death")
 line_chart = plotting.make_region_since_chart(
     base,
     'dead',
     'sinceDay0',
     'region',
     'Days since 10th death',
     'Cumulative deaths',
     'Deaths',
     'Region'
 ).properties(
     width=450,
     height=450
 )
 rule_chart = plotting.make_rule_chart(
     start_case=10,
     max_case=50000,
     max_days=40,
     pos_day=(12,50000),
     pos_3days=(35,20000),
     pos_week=(40,100))
 # add the total deaths for italy to the chart
 since_10th_death_all = helper.make_since_df(df_national, start_case=10, column='dead', region_column='country_code')
 since_10th_death_all['country_label'] = 'Italy'
 italy = alt.Chart(since_10th_death_all).mark_line(color='steelblue').encode(x='sinceDay0', y='dead')
 italy += italy.mark_text(dy=-10).encode(text='country_label').transform_filter(f"datum['sinceDay0'] == {len(since_10th_death_all)-1}")
 line_chart + rule_chart + italy
 ```
 %% Output
     alt.LayerChart(...)
 %% Cell type:code id: tags:
 ``` python
 ```