In [1]:

    

# Import all necessary libraries
import pandas     as pd   # The data processing library
import matplotlib as plot # The visualization library
import json               # Library for encoding and decoding JSON data


    

In [2]:

    

# Load GeoJSON directly in a data frame
bad_geo_df = pd.read_json("../datasets/connectivity_2015_09.geojson")
bad_geo_df.head()


    

    
Out[2]:






  

    

      

      
features
      
type
    
  
  

    

      
0
      
{u'geometry': {u'type': u'LineString', u'coord...
      
FeatureCollection
    
    

      
1
      
{u'geometry': {u'type': u'LineString', u'coord...
      
FeatureCollection
    
    

      
2
      
{u'geometry': {u'type': u'LineString', u'coord...
      
FeatureCollection
    
    

      
3
      
{u'geometry': {u'type': u'LineString', u'coord...
      
FeatureCollection
    
    

      
4
      
{u'geometry': {u'type': u'LineString', u'coord...
      
FeatureCollection
    
  

In [3]:

    

# Load JSON and normalize nested JSON
with open("../datasets/connectivity_2015_09.geojson") as f:
    geo_data = json.load(f)

tmp_df = pd.DataFrame(geo_data)
geo_df = pd.io.json.json_normalize(tmp_df['features'])
geo_df.head()


    

    
Out[3]:






  

    

      

      
geometry.coordinates
      
geometry.type
      
properties.all,no4g_measurements
      
properties.all,no4g_stability
      
properties.all_measurements
      
properties.all_stability
      
properties.e-plus,no4g_measurements
      
properties.e-plus,no4g_stability
      
properties.e-plus_measurements
      
properties.e-plus_stability
      
...
      
properties.o2_stability
      
properties.t-mobile,no4g_measurements
      
properties.t-mobile,no4g_stability
      
properties.t-mobile_measurements
      
properties.t-mobile_stability
      
properties.vodafone,no4g_measurements
      
properties.vodafone,no4g_stability
      
properties.vodafone_measurements
      
properties.vodafone_stability
      
type
    
  
  

    

      
0
      
[[11.06535, 49.44201], [11.07071, 49.4427], [1...
      
LineString
      
567
      
0.974
      
1157
      
0.961
      
94
      
0.981
      
112
      
0.981
      
...
      
0.979
      
188
      
0.971
      
437
      
0.929
      
205
      
0.980
      
364
      
0.981
      
Feature
    
    

      
1
      
[[11.0718650022, 49.4429816674], [11.07228, 49...
      
LineString
      
1154
      
0.973
      
2161
      
0.954
      
150
      
0.974
      
185
      
0.974
      
...
      
0.968
      
418
      
0.976
      
896
      
0.924
      
448
      
0.979
      
751
      
0.977
      
Feature
    
    

      
2
      
[[11.0779766624, 49.4447023689], [11.07897, 49...
      
LineString
      
1569
      
0.973
      
2818
      
0.947
      
160
      
0.968
      
208
      
0.967
      
...
      
0.959
      
618
      
0.980
      
1274
      
0.918
      
621
      
0.978
      
991
      
0.975
      
Feature
    
    

      
3
      
[[11.0844614667, 49.4457747006], [11.0882, 49....
      
LineString
      
1289
      
0.972
      
2370
      
0.934
      
112
      
0.960
      
155
      
0.959
      
...
      
0.957
      
497
      
0.981
      
1131
      
0.895
      
534
      
0.977
      
814
      
0.975
      
Feature
    
    

      
4
      
[[11.0908624199, 49.4470433591], [11.09168, 49...
      
LineString
      
594
      
0.974
      
1282
      
0.888
      
48
      
0.961
      
71
      
0.964
      
...
      
0.964
      
213
      
0.980
      
697
      
0.817
      
248
      
0.980
      
371
      
0.978
      
Feature
    
  

5 rows × 23 columns

In [4]:

    

# List the column headers
list(geo_df.columns.values)


    

    
Out[4]:





['geometry.coordinates',
 'geometry.type',
 'properties.all,no4g_measurements',
 'properties.all,no4g_stability',
 'properties.all_measurements',
 'properties.all_stability',
 'properties.e-plus,no4g_measurements',
 'properties.e-plus,no4g_stability',
 'properties.e-plus_measurements',
 'properties.e-plus_stability',
 'properties.o2,no4g_measurements',
 'properties.o2,no4g_stability',
 'properties.o2_measurements',
 'properties.o2_stability',
 'properties.t-mobile,no4g_measurements',
 'properties.t-mobile,no4g_stability',
 'properties.t-mobile_measurements',
 'properties.t-mobile_stability',
 'properties.vodafone,no4g_measurements',
 'properties.vodafone,no4g_stability',
 'properties.vodafone_measurements',
 'properties.vodafone_stability',
 u'type']

In [10]:

    

# Let's examine the coordinate
print geo_df['geometry.coordinates'][0], "\n"
print geo_df['geometry.coordinates'][1], "\n"
print geo_df['geometry.coordinates'][2], "\n"


    

    




[[11.06535, 49.44201], [11.07071, 49.4427], [11.07126, 49.4428], [11.07177, 49.44295], [11.071865002158022, 49.44298166738601]] 

[[11.071865002158022, 49.44298166738601], [11.07228, 49.44312], [11.07278, 49.44331], [11.07325, 49.44349], [11.07425, 49.44383], [11.0747, 49.444], [11.07569, 49.44429], [11.0762, 49.44441], [11.07675, 49.4445], [11.07731, 49.44461], [11.077976662356802, 49.44470236888076]] 

[[11.077976662356802, 49.44470236888076], [11.07897, 49.44484], [11.08006, 49.44496], [11.08115, 49.44511], [11.08223, 49.4453], [11.084461466732163, 49.44577470062812]] 

In [28]:

    

# We can import a module wherever we want
import matplotlib as mpl
import matplotlib.pyplot as plt
from matplotlib.patches import Polygon
from matplotlib.collections import PolyCollection
import numpy as np
%matplotlib inline

# Initialize a collection of Polygons
fig, ax = plt.subplots()

coll = PolyCollection(geo_df['geometry.coordinates'], 
                      cmap=mpl.cm.jet,
                      edgecolors='none',
                      zorder=2,
                      color='blue')

ax.add_collection(coll)
ax.autoscale_view()
plt.show()


    

In [ ]: