def serialize_as_altair(
topo_object,
mesh=True,
color=None,
tooltip=True,
projection="identity",
objectname="data",
geo_interface=False,
):
import altair as alt
# create a mesh visualization
if geo_interface and mesh:
# chart object
chart = (
alt.Chart(topo_object, width=300)
.mark_geoshape(filled=False)
.project(type=projection, reflectY=True)
)
# create a mesh visualization
elif mesh and color is None:
data = alt.InlineData(
values=topo_object, format=alt.DataFormat(mesh=objectname, type="topojson")
)
chart = (
alt.Chart(data, width=300)
.mark_geoshape(filled=False)
.project(type=projection, reflectY=True)
)
# creating a chloropleth visualisation
elif color is not None:
data = alt.InlineData(
values=topo_object,
format=alt.DataFormat(feature=objectname, type="topojson"),
)
if tooltip is True:
tooltip = [color]
chart = (
alt.Chart(data, width=300)
.mark_geoshape()
.encode(
color=alt.Color(color, legend=alt.Legend(columns=2)), tooltip=tooltip
)
.project(type=projection, reflectY=True)
)
return chart
def convert_gfp_to_alt(gdf: geopandas.geodataframe) -> alt.InlineData:
"""
Converts a geopandas geoDataFrame into a Altair InlineData instance so it can be converted into a chart.
Args:
gdf (geopandas.geodataframe): Input geodataframe
Returns:
alt.InlineData: An instance of Altair's InlineData class.
"""
data = alt.InlineData(
values=gdf.to_json(), # geopandas to geojson string
# root object type is "FeatureCollection" but we need its features
format=alt.DataFormat(property="features", type="json"),
)
return data
def load_district_arrests():
# url for district arrests
crime_url = "https://data.cityofchicago.org/resource/ijzp-q8t2.json?$select=year,district,count(district),sum(case(arrest='1',1,true,0))&$group=year,district"
#crime_url = "https://data.cityofchicago.org/resource/ijzp-q8t2.json?$select=year,district,count(district)&$group=year,district"
df_district_count = pd.read_json(crime_url)
df_district_count['district'] = df_district_count['district'].dropna().astype(int).astype(str)
url_geojson = "https://data.cityofchicago.org/resource/24zt-jpfn.geojson"
data_geojson_remote = alt.Data(url=url_geojson, format=alt.DataFormat(property='the_geom',type='json'))
stations = pd.read_json("https://data.cityofchicago.org/resource/z8bn-74gv.json?$select=district,latitude,longitude")
stations = stations[stations['district'] != 'Headquarters']
#st.write(type(stations['district']))
#st.write(type(df_district_count['district']))
data = pd.merge(stations, df_district_count, on='district')
data.columns = ['district','latitude','longitude','year','count_district','arrests']
data['arrest_rate'] = data['arrests']/data['count_district']
return data
def country_codes():
cc = coco.CountryConverter()
df = pd.read_csv(
'https://query.data.world/s/atfqu76bmkmwiranfkzfiqqdnvpzle',
encoding="ISO-8859-1")
url_geojson = "https://query.data.world/s/ziovq5tz5zcqvizgghygjucscii32t"
countries = alt.Data(url=url_geojson,
format=alt.DataFormat(property="features",
type="json"))
unique = pd.DataFrame(df["Country"].unique())
unique = unique.rename(columns={0: "Country"})
unique['iso_a3'] = unique["Country"].apply(
lambda x: coco.convert(names=x, to="ISO3"))
df = pd.merge(df,
unique,
how="left",
left_on="Country",
right_on="Country")
return df, countries, unique
import altair as alt
from vega_datasets import data
alt.renderers.enable('altair_viewer')
movies = alt.UrlData(data.movies.url,
format=alt.DataFormat(parse={"Release_Date": "date"}))
ratings = ['G', 'NC-17', 'PG', 'PG-13', 'R']
genres = [
'Action', 'Adventure', 'Black Comedy', 'Comedy', 'Concert/Performance',
'Documentary', 'Drama', 'Horror', 'Musical', 'Romantic Comedy',
'Thriller/Suspense', 'Western'
]
base = alt.Chart(movies, width=200,
height=200).mark_point(filled=True).transform_calculate(
Rounded_IMDB_Rating="floor(datum.IMDB_Rating)",
Hundred_Million_Production=
"datum.Production_Budget > 100000000.0 ? 100 : 10",
Release_Year="year(datum.Release_Date)").transform_filter(
alt.datum.IMDB_Rating > 0).transform_filter(
alt.FieldOneOfPredicate(
field='MPAA_Rating',
oneOf=ratings)).encode(x=alt.X(
'Worldwide_Gross:Q',
scale=alt.Scale(domain=(100000, 10**9),
clamp=True)),
y='IMDB_Rating:Q',
tooltip="Title:N")
# A slider filter
year_slider = alt.binding_range(min=1969, max=2018, step=1)
slider_selection = alt.selection_single(bind=year_slider,
#st.write(type(stations['district']))
#st.write(type(df_district_count['district']))
data = pd.merge(stations, df_district_count, on='district')
data.columns = ['district','latitude','longitude','year','count_district','arrests']
data['arrest_rate'] = data['arrests']/data['count_district']
return data
data = load_district_arrests()
#st.write(data)
#### new stuff from vivian ####
url_geojson = "https://data.cityofchicago.org/resource/24zt-jpfn.geojson"
data_geojson_remote = alt.Data(url=url_geojson, format=alt.DataFormat(property='the_geom',type='json'))
stations = pd.read_json("https://data.cityofchicago.org/resource/z8bn-74gv.json")
stations = stations[stations['district'] != 'Headquarters']
# print(stations)
#st.write(stations)
#url_geojson = "https://data.cityofchicago.org/resource/24zt-jpfn.geojson"
data_geojson_remote = alt.Data(url=url_geojson, format=alt.DataFormat(property='the_geom',type='json'))
# chart object
district_map = alt.Chart(data_geojson_remote).mark_geoshape(
fill='lightgray',
stroke='white'
corr = tracts_gdf_selected[var_demographic2].corr()
corr.style.background_gradient(cmap='coolwarm')
# It turns out that racial composition (pct_white) is substantially related to education level (pct_bachelor). It's surprisingly that mdeian household income is statistically associated with population, that is, tracts with larger number of population exhibit higher income level.
# ### Spatial Patterns
# In[34]:
import altair as alt
alt.renderers.enable('notebook')
# In[44]:
tracts_alt = alt.InlineData(values=tracts_gdf_selected.dropna().to_json(),
format=alt.DataFormat(property='features',
type='json'))
alt.Chart(tracts_alt).mark_geoshape(stroke='population', ).properties(
width=500,
height=400,
projection={
"type": 'mercator'
},
).encode(tooltip=['properties.population:Q', 'properties.NAME:N'],
color='properties.population:Q')
# In[45]:
alt.Chart(tracts_alt).mark_geoshape(stroke='pct_male', ).properties(
width=500,
height=400,
def generate_regions_map(data: pd.DataFrame,
date,
feature: str,
title: str,
width: int = 600,
height: int = 600,
log_scale: bool = True) -> alt.Chart:
fechas = data['fecha'].apply(lambda x: x.date())
data = data[(fechas == date)]
url_geojson = 'https://raw.githubusercontent.com/rubens36/covid19/master/maps/regiones.geojson'
regions_shape = alt.Data(url=url_geojson,
format=alt.DataFormat(property='features',
type='json'))
chart_data = data[(data[feature] > 0)][[feature, 'id']]
base_chart = (alt.Chart(
regions_shape,
title='Ubicación de los casos confirmados por región').mark_geoshape(
stroke='black', strokeWidth=0.5, color='white').encode(
tooltip=[alt.Tooltip('properties.Region:N', title='Región')]))
scale = alt.Scale(type='log', scheme='teals') if log_scale else alt.Scale(
type='linear', scheme='teals')
color_chart = (alt.Chart(regions_shape).mark_geoshape(
stroke='black', strokeWidth=0.5).encode(
color=alt.Color(f"{feature}:Q",
title=(get_title(feature)),
scale=scale),
tooltip=[
alt.Tooltip('properties.Region:N', title='Región'),
alt.Tooltip(f"{feature}:Q", title=(get_title(feature)))
]).transform_lookup('properties.id',
from_=alt.LookupData(data=chart_data,
key='id',
fields=[feature])))
single = alt.selection_single(on='mouseover')
circle_chart = (alt.Chart(regions_shape).mark_circle(
opacity=0.4, color='red').encode(
longitude='properties.centroid_lon:Q',
latitude='properties.centroid_lat:Q',
size=(alt.condition(
single,
alt.Size('infectados:Q',
scale=alt.Scale(range=[50, 4000]),
legend=None), alt.value(0))),
tooltip=[
alt.Tooltip('properties.Region:N', title='Región'),
alt.Tooltip(f"{feature}:Q", title=(get_title(feature)))
]).transform_lookup('properties.id',
from_=alt.LookupData(
data=chart_data,
key='id',
fields=['infectados'
])).properties(selection=single))
final_chart = ((base_chart + color_chart + circle_chart).configure_view(
strokeWidth=0).properties(width=width, height=height))
return final_chart
color_continuous_scale=px.colors.sequential.Viridis,
height=1200,
)
fig.update_geos(fitbounds="locations", visible=False)
fig.update_layout(margin={"r": 0, "t": 0, "l": 0, "b": 0})
fig
fig.write_html("plotly_choropleth.html")
# +
# aug_df.crs = {"init": "epsg:27700"}
# aug_df = aug_df.to_crs({"init": "epsg:4326"})
# define inline geojson data object
data_geojson = alt.InlineData(
values=aug_df.to_json(), format=alt.DataFormat(property="features", type="json")
)
# chart object
poverty_chart = (
alt.Chart(data_geojson)
.mark_geoshape()
.encode(
color=alt.Color("properties.Taux de pauvreté:Q", legend=alt.Legend(columns=2),),
tooltip=[
"properties.Délégation:N",
"properties.Gouvernorat:N",
"properties.Taux de pauvreté:Q",
],
)
.properties(width=500, height=700)
color=alt.condition(
selector_map,
alt.Color("properties.region:N",
scale=alt.Scale(scheme="category20")),
alt.value("lightgray"),
),
tooltip=[
alt.Tooltip("properties.name:N", title="County"),
alt.Tooltip("properties.region:N", title="Region"),
],
).add_selection(selector_map).properties(width=500,
height=800))
bar_chart = (alt.Chart(
alt.Data(url=data_url,
format=alt.DataFormat(
property="features", type="json"))).transform_fold(
["properties." + c for c in columns],
as_=["metric", "value"]).mark_bar().encode(
x=alt.X("metric:N"),
y=alt.Y("value:Q", axis=alt.Axis(format="%")),
color="properties.region:N",
).add_selection(selector_map).transform_filter(
selector_map).properties(width=200, height=200))
chart = CA_map | bar_chart
if __name__ == "__main__":
docs_dir = Path(__file__).parent.parent.absolute()
with open(docs_dir / "docs" / "_vega_out.html", "w") as f:
