Put your data on map with ease

Anastazie and Filip Sedlakovi

sedlakovi.org

Workshop programme

  • Check installation
  • Introduction
  • Data cleaning
  • Plotly
  • Folium
  • Tuning maps

Who are we

  • Anastazie
    • Data Scientist
    • Background: gentics and forensics
  • Filip
    • DevOps
    • Background: chemoinformatics
  • Anastazie and Filip
    • Organizing programming courses
    • Working/consulting on data ananlysis/ devops projects
    • Visit our courses

Usecase of this workshop

You are data scientist who uses Python almost daily for data cleaning and modelling. Sometimes you need to visualize results of your work. This time you need to plot your data on the map. You do not do it routinely (and you don't have background in geography), but you want to create quickly nice interactive visualization.

Type of visualization on maps

Scatter map

  • Displays geographical data as data points on a map
    • E.g. list of US ZIP Code Tabulation Areas used by Census Bureau

Bubble maps

  • Circles are displayed over a designated geographical region
  • Area of the circle proportional to its value in the dataset
    • E.g. total alcohol consumption

Choropleth maps

  • Display divided geographical areas or regions that are coloured, shaded or patterned in relation to a data variable
    • E.g. total alcohol consumption

Import numpy and pandas

In [1]:
import numpy as np
import pandas as pd

Plotly

  • Interactive, open-source, and browser-based graphing library for Python
  • Each graph comes with built-in set of tools
    • Zooming, selecting area, saving png
  • For choropleth maps, if you want other areas than coutries/US states, use Mapbox
  • Is not rendered by github
  • Gallery
from plotly.offline import iplot, init_notebook_mode, plot
import plotly.graph_objs as go
init_notebook_mode()

x = np.random.randn(500)
data = [go.Histogram(x=x)]
layout = go.Layout(title='Example histogram', xaxis=dict(title='Random data'))
fig = go.Figure(data=data, layout=layout)

iplot(fig)

  • plotly.graph_objs library contains graphical objects
    • E.g. Histogram, Bar, Scatter, Choropleth
    • Also Figure and Layout
    • You can also use syntax without graphical objects using dict()
  • plotly.offline enables showing plots without sending data to plotly
  • init_notebook_mode()
    • Will make plots visible in Jupyter notebook
  • Figure cntains data to plot (must be a list) and layout 
# dict() syntax
x = np.random.randn(500)
data = [dict(type='histogram', x=x)]
layout = dict(title='Example histogram', xaxis=dict(title='Random data'))
fig = dict(data=data, layout=layout)

iplot(fig)

Maps in Plotly

Link to original tutorial

df = pd.read_csv('https://raw.githubusercontent.com/plotly/datasets/master/2014_world_gdp_with_codes.csv')
data = [go.Choropleth(
    locations = df['CODE'],
    z = df['GDP (BILLIONS)'],
    text = df['COUNTRY'],
    colorscale = 'Viridis',
    autocolorscale = False,
    reversescale = True,
    marker = go.choropleth.Marker(
        line = go.choropleth.marker.Line(
            color = 'rgb(180,180,180)',
            width = 0.5
        ))
)]
layout = go.Layout(
    title = go.layout.Title(
        text = '2014 Global GDP'
    ),
    geo = go.layout.Geo(
        showframe = False,
        showcoastlines = False,
        projection = go.layout.geo.Projection(
            type = 'equirectangular'
        )
    )
)
fig = go.Figure(data = data, layout = layout)
iplot(fig)

Folium

  • Library devoted specifically to create interactive visualizations on map
  • Python wrapper for JavaScript mapping library Leaflet.js
  • Each map comes with possibility of zooming and moving area
  • If you want English names on the whole map - use Mapbox
  • Is not rendered by github
  • Gallery

Link to original tutorial

import folium
file_url = 'http://www2.census.gov/geo/docs/maps-data/data/gazetteer/2016_Gazetteer/2016_Gaz_zcta_national.zip'
df = pd.read_csv(file_url, sep='\t', dtype={'GEOID' : object})
#some column names have some extra padding
df.columns = df.columns.str.strip()
subset_df = df.sample(n=500)
example_map = folium.Map(
    location=[subset_df['INTPTLAT'].mean(), subset_df['INTPTLONG'].mean()], zoom_start=4
)
for row in subset_df.itertuples():
    example_map.add_child(folium.Marker(location=[row.INTPTLAT,row.INTPTLONG],popup=row.GEOID))
example_map

  • In the beginning, we need to create Map object
  • Icons are added to Map by creating Marker object for each point
    • Markers can be aggregated by adding them to MarkerCluster instead of Map directly
  • Layers are added to Map using add_child method
  • Clickable layers can be added to Map using folium.LayerControl().add_to(example_map)

Link to original tutorial

import folium.plugins
file_url = 'http://www2.census.gov/geo/docs/maps-data/data/gazetteer/2016_Gazetteer/2016_Gaz_zcta_national.zip'
df = pd.read_csv(file_url, sep='\t', dtype={'GEOID' : object})
#some column names have some extra padding
df.columns = df.columns.str.strip()
subset_df = df.sample(n=500)
mc = folium.plugins.MarkerCluster()
example_map = folium.Map(
    location=[subset_df['INTPTLAT'].mean(), subset_df['INTPTLONG'].mean()], zoom_start=4
)
for row in subset_df.itertuples():
    mc.add_child(folium.Marker(location=[row.INTPTLAT,row.INTPTLONG],
           popup=row.GEOID))
example_map.add_child(mc)
example_map

Link to original tutorial

url = 'https://raw.githubusercontent.com/python-visualization/folium/master/examples/data'
state_geo = f'{url}/us-states.json'
state_unemployment = f'{url}/US_Unemployment_Oct2012.csv'
state_data = pd.read_csv(state_unemployment)
example_map = folium.Map(location=[48, -102], zoom_start=3)
folium.Choropleth(
    geo_data=state_geo,
    name='choropleth',
    data=state_data,
    columns=['State', 'Unemployment'],
    key_on='feature.id',
    fill_color='YlGn',
    fill_opacity=0.7,
    line_opacity=0.2,
    legend_name='Unemployment Rate (%)'
).add_to(example_map)
folium.LayerControl().add_to(example_map)
example_map