Open Data: Germany Maps Viz

First-Digit-Postalcodes Areas

Next, let us plot different regions corresponding to the first digit of each postal code.

# Create feature.
plz_shape_df = plz_shape_df \
    .assign(first_dig_plz = lambda x: x['plz'].str.slice(start=0, stop=1))

fig, ax = plt.subplots()

plz_shape_df.plot(
    ax=ax, 
    column='first_dig_plz', 
    categorical=True, 
    legend=True, 
    legend_kwds={'title':'First Digit', 'loc':'lower right'},
    cmap='tab20',
    alpha=0.9
)

for c in top_cities.keys():

    ax.text(
        x=top_cities[c][0], 
        y=top_cities[c][1] + 0.08, 
        s=c, 
        fontsize=12,
        ha='center', 
    )

    ax.plot(
        top_cities[c][0], 
        top_cities[c][1], 
        marker='o',
        c='black', 
        alpha=0.5
    )

ax.set(
    title='Germany First-Digit-Postal Codes Areas', 
    aspect=1.3,
    facecolor='white'
);

Bundesland Map

Let us now map each postal code to the corresponding region:

plz_region_df = pd.read_csv(
    '../Data/zuordnung_plz_ort.csv', 
    sep=',', 
    dtype={'plz': str}
)

plz_region_df.drop('osm_id', axis=1, inplace=True)

plz_region_df.head()

# Merge data.
germany_df = pd.merge(
    left=plz_shape_df, 
    right=plz_region_df, 
    on='plz',
    how='inner'
)

germany_df.drop(['note'], axis=1, inplace=True)

germany_df.head()

Generate Bundesland map:

fig, ax = plt.subplots()

germany_df.plot(
    ax=ax, 
    column='bundesland', 
    categorical=True, 
    legend=True, 
    legend_kwds={'title':'Bundesland', 'bbox_to_anchor': (1.35, 0.8)},
    cmap='tab20',
    alpha=0.9
)

for c in top_cities.keys():

    ax.text(
        x=top_cities[c][0], 
        y=top_cities[c][1] + 0.08, 
        s=c, 
        fontsize=12,
        ha='center', 
    )

    ax.plot(
        top_cities[c][0], 
        top_cities[c][1], 
        marker='o',
        c='black', 
        alpha=0.5
    )

ax.set(
    title='Germany - Bundesländer', 
    aspect=1.3, 
    facecolor='white'
);

Number of Inhabitants

Now we include the number of inhabitants per postal code:

plz_einwohner_df = pd.read_csv(
    '../Data/plz_einwohner.csv', 
    sep=',', 
    dtype={'plz': str, 'einwohner': int}
)

plz_einwohner_df.head()

# Merge data.
germany_df = pd.merge(
    left=germany_df, 
    right=plz_einwohner_df, 
    on='plz',
    how='left'
)

germany_df.head()

Generate map:

fig, ax = plt.subplots()

germany_df.plot(
    ax=ax, 
    column='einwohner', 
    categorical=False, 
    legend=True, 
    cmap='autumn_r',
    alpha=0.8
)

for c in top_cities.keys():

    ax.text(
        x=top_cities[c][0], 
        y=top_cities[c][1] + 0.08, 
        s=c, 
        fontsize=12,
        ha='center', 
    )

    ax.plot(
        top_cities[c][0], 
        top_cities[c][1], 
        marker='o',
        c='black', 
        alpha=0.5
    )
    
ax.set(
    title='Germany: Number of Inhabitants per Postal Code', 
    aspect=1.3, 
    facecolor='lightblue'
);

Berlin Neighbourhoods

Fortunately, the website http://insideairbnb.com/get-the-data.html, containing AirBnB data for many cities (which is definitely worth investigatinig!), has a convinient data set neighbourhoods.geojson which maps Berlin’s area to neighbourhoods:

berlin_neighbourhoods_df = gpd.read_file('../Data/neighbourhoods.geojson')

berlin_neighbourhoods_df = berlin_neighbourhoods_df \
    [~ berlin_neighbourhoods_df['neighbourhood_group'].isnull()]

berlin_neighbourhoods_df.head()

fig, ax = plt.subplots()

berlin_df.plot(
    ax=ax, 
    alpha=0.2
)

berlin_neighbourhoods_df.plot(
    ax=ax, 
    column='neighbourhood_group',
    categorical=True, 
    legend=True, 
    legend_kwds={'title': 'Neighbourhood', 'loc': 'upper right'},
    cmap='tab20', 
    edgecolor='black'
)

ax.set(
    title='Berlin Neighbourhoods', 
    aspect=1.3
);

Here the divisions correspond to Neighbourhood \(\subset\) Neighbourhood Group.

Selected Locations in Berlin

Sometimes it is useful to include well-known locations on the maps so that the user can identify them and understand the distances and scales. One way to do it is to manualy input the latitude and longitude of each point (as above). This of course can be time consuming and prone to errors. As expected, there is a library which can fetch this type of data automatically, namely geopy.

Here is a simple example:

from geopy import Nominatim

locator = Nominatim(user_agent='myGeocoder')

location = locator.geocode('Humboldt Universität zu Berlin')

print(location)

Humboldt-Universität zu Berlin, Dorotheenstraße, Spandauer Vorstadt, Mitte, Berlin, 10117, Deutschland

Let us write a function to get the latitude and longitude coordinates:

def lat_lng_from_string_loc(x):
    
    locator = Nominatim(user_agent='myGeocoder')

    location = locator.geocode(x)
    
    if location is None:
        None
    else:
        return location.longitude, location.latitude

# Define some well-known Berlin locations.
berlin_locations = [
    'Alexander Platz', 
    'Zoo Berlin', 
    'Berlin Tegel', 
    'Berlin Schönefeld',
    'Berlin Adlershof',
    'Olympia Stadium Berlin',
    'Berlin Südkreuz', 
    'Frei Universität Berlin',
    'Mauerpark', 
    'Treptower Park',
]

# Get geodata.
berlin_locations_geo = {
    x: lat_lng_from_string_loc(x) 
    for x in berlin_locations 
}

# Remove None.
berlin_locations_geo = {
    k: v 
    for k, v in berlin_locations_geo.items()
    if v is not None
}

Let us see the resulting map:

berlin_df = germany_df.query('ort == "Berlin"')

fig, ax = plt.subplots()

berlin_df.plot(
    ax=ax, 
    color='orange', 
    alpha=0.8
)

for c in berlin_locations_geo.keys():

    ax.text(
        x=berlin_locations_geo[c][0], 
        y=berlin_locations_geo[c][1] + 0.005, 
        s=c, 
        fontsize=12,
        ha='center', 
    )

    ax.plot(
        berlin_locations_geo[c][0], 
        berlin_locations_geo[c][1], 
        marker='o',
        c='black', 
        alpha=0.5
    )

ax.set(
    title='Berlin - Some Relevant Locations', 
    aspect=1.3,
    facecolor='lightblue'
);

Christmas Markets

Let us enrich the maps by including other type of information. There is a great resource for publicly available data for Berlin: Berlin Open Data. Among many interesting datasets, I found one on the Christmas markets around the city (which are really fun to visit!) here. You can acces the data via a public API. Let us use the requests module to do this:

import requests

# GET request.
response = requests.get(
    'https://www.berlin.de/sen/web/service/maerkte-feste/weihnachtsmaerkte/index.php/index/all.json?q='
)

response_json = response.json()

Convert to pandas dataframe.

berlin_maerkte_raw_df = pd.DataFrame(response_json['index'])

We do not have a postal code feature, but we can create one by extracting it from the plz_ort column.

berlin_maerkte_df = berlin_maerkte_raw_df[['name', 'bezirk', 'plz_ort', 'lat', 'lng']]

berlin_maerkte_df = berlin_maerkte_df \
    .query('lat != ""') \
    .assign(plz = lambda x: x['plz_ort'].str.split(' ').apply(lambda x: x[0]).astype(str)) \
    .drop('plz_ort', axis=1)

# Convert to float.
berlin_maerkte_df['lat'] = berlin_maerkte_df['lat'].str.replace(',', '.').astype(float)
berlin_maerkte_df['lng'] = berlin_maerkte_df['lng'].str.replace(',', '.').astype(float)

berlin_maerkte_df.head()

Let us plot the Christmas markets locations:

fig, ax = plt.subplots()

berlin_df.plot(ax=ax, color= 'green', alpha=0.4)

for c in berlin_locations_geo.keys():

    ax.text(
        x=berlin_locations_geo[c][0], 
        y=berlin_locations_geo[c][1] + 0.005, 
        s=c, 
        fontsize=12,
        ha='center', 
    )

    ax.plot(
        berlin_locations_geo[c][0], 
        berlin_locations_geo[c][1], 
        marker='o',
        c='black', 
        alpha=0.5
    )

berlin_maerkte_df.plot(
    kind='scatter', 
    x='lng', 
    y='lat', 
    c='r', 
    marker='*',
    s=50,
    label='Christmas Market',  
    ax=ax
)

ax.set(
    title='Berlin Christmas Markets (2019)', 
    aspect=1.3, 
    facecolor='white'
);