In [1]:

import pandas
import matplotlib.pyplot as plt

In [2]:

cities = pandas.read_csv('US_cities.csv')
import geopandas
from geopandas import GeoDataFrame
from shapely.geometry import Point

geometry = [Point(xy) for xy in zip(cities.Longitude, cities.Latitude)]
cities = cities.drop(['Longitude', 'Latitude'], axis = 1)
crs = {'init':'epsg:4326'}
cities = GeoDataFrame(cities, crs=crs, geometry=geometry)
# cities = cities.to_crs({'init': 'epsg:3395'})
cities.head()

Out[2]:

	id	Country	City	AccentCity	Region	Population	geometry
0	2907755	us	alabaster	Alabaster	AL	26738.0	POINT (-86.81638890000001 33.2441667)
1	2907759	us	albertville	Albertville	AL	18368.0	POINT (-86.20888890000001 34.2675)
2	2907765	us	alexander city	Alexander City	AL	14993.0	POINT (-85.95388890000002 32.9438889)
3	2907804	us	anniston	Anniston	AL	23423.0	POINT (-85.8316667 33.6597222)
4	2907848	us	athens	Athens	AL	20470.0	POINT (-86.9716667 34.8027778)

In [3]:

cafes = geopandas.read_file('cafes_us_OSM.geojson')
cafes.head()

Out[3]:

	id	@id	amenity	building	cuisine	name	operator	wheelchair	website	addr:city	...	karaoke	gambling	reservation	tourism:disused	indoor:level	landuse	diet:raw	name:it	amenity:ice_cream	geometry
0	node/31327678	node/31327678	cafe	yes	None	None	None	None	None	None	...	None	None	None	None	None	None	None	None	None	POINT (-122.330085 47.5454108)
1	node/32153673	node/32153673	cafe	None	coffee_shop	Siuslaw Coffee Roasters	None	None	None	None	...	None	None	None	None	None	None	None	None	None	POINT (-124.1081239 43.9662488)
2	node/33979948	node/33979948	cafe	None	None	Country Style	MTY Group	yes	None	None	...	None	None	None	None	None	None	None	None	None	POINT (-76.4835513 44.2569218)
3	node/34084322	node/34084322	cafe	None	coffee_shop	Tim Hortons	Restaurant Brands International Inc.	None	http://www.timhortons.com/	None	...	None	None	None	None	None	None	None	None	None	POINT (-76.5066058 44.2598381)
4	node/35332415	node/35332415	cafe	None	coffee_shop	Starbucks	None	None	None	None	...	None	None	None	None	None	None	None	None	None	POINT (-76.9346255 40.2387067)

5 rows × 400 columns

In [4]:

cafes.sample().T.dropna()

Out[4]:

	6889
id	node/2157948579
@id	node/2157948579
amenity	cafe
cuisine	bubble_tea
name	Verde Tea Cafe
addr:city	Mountain View
addr:housenumber	852
addr:street	Villa Street
geometry	POINT (-122.0794931 37.3941483)

In [5]:

cafes['count'] = 1
cafes[['cuisine', 'count']].groupby('cuisine').sum().sort_values('count', ascending=False)[:10]

Out[5]:

	count
cuisine
coffee_shop	6166
donut	625
ice_cream	290
sandwich	284
american	220
coffee	78
cafe	73
regional	71
bagel	64
tea	52

In [6]:

sorted(list(cafes.columns))[:20]

Out[6]:

['@id',
 'Door',
 'FIXME',
 'Notes',
 'access',
 'addr',
 'addr.source:street',
 'addr:city',
 'addr:country',
 'addr:county',
 'addr:direction',
 'addr:door',
 'addr:flats',
 'addr:floor',
 'addr:full',
 'addr:house',
 'addr:housename',
 'addr:housenumber',
 'addr:interpolation',
 'addr:place']

In [7]:

cafes.isnull().sum() / len(cafes)

Out[7]:

id                          0.000000
@id                         0.000000
amenity                     0.000000
building                    0.998607
cuisine                     0.552107
name                        0.030798
operator                    0.970932
wheelchair                  0.950800
website                     0.814443
addr:city                   0.763946
addr:country                0.961514
addr:housenumber            0.665257
addr:province               0.971412
addr:street                 0.657954
internet_access             0.852784
opening_hours               0.866766
payment:amex                0.985154
payment:coins               0.984817
payment:notes               0.984817
recycling:coffee_cups       0.999856
created_by                  0.995628
addr:postcode               0.764186
description                 0.992024
hst_number                  0.999760
self_checkout               0.999952
capacity                    0.998847
smoking                     0.948734
wifi                        0.971748
wheelchair:description      0.998462
source                      0.943689
                              ...   
name:nl                     0.999952
source:phone                0.999952
payment:cheque              0.999904
dessert:type                0.999952
payment:discover            0.999952
payment:paypal              0.999952
cat                         0.999952
name:zh_pinyin              0.999952
service:bicycle:cleaning    0.999952
service:bicycle:parts       0.999952
service:bicycle:pump        0.999952
Door                        0.999952
automatic_door              0.999952
addr:street:prefix          0.999952
black_owned                 0.999952
ref:miaaddr                 0.999952
service                     0.999952
name:etymology:wikidata     0.999952
name:pt                     0.999904
karaoke                     0.999952
gambling                    0.999952
reservation                 0.999952
tourism:disused             0.999952
indoor:level                0.999856
landuse                     0.999952
diet:raw                    0.999952
name:it                     0.999952
amenity:ice_cream           0.999952
geometry                    0.000000
count                       0.000000
dtype: float64

In [8]:

cafes = cafes.T[cafes.isnull().sum() / len(cafes) < 0.6].T
cafes.sample(5)

Out[8]:

	id	@id	amenity	cuisine	name	geometry	count
12245	node/3367217305	node/3367217305	cafe	None	Tropical Smoothie Cafe	POINT (-113.5616411 37.1121392)	1
9133	node/2605729050	node/2605729050	cafe	coffee_shop	Dunkin Donuts (Summer Only)	POINT (-70.3747085 43.515823)	1
15173	node/4045366894	node/4045366894	cafe	donut	Dunkin' Donuts	POINT (-70.5956206 41.566752)	1
17698	node/4514634690	node/4514634690	cafe	coffee_shop	Starbucks	POINT (-78.6799591 35.8571021)	1
7638	node/2332676900	node/2332676900	cafe	None	Starbucks	POINT (-111.8926801 40.588471)	1

In [9]:

del cafes['id']
del cafes['@id']
cafes.head()

Out[9]:

	amenity	cuisine	name	geometry	count
0	cafe	None	None	POINT (-122.330085 47.5454108)	1
1	cafe	coffee_shop	Siuslaw Coffee Roasters	POINT (-124.1081239 43.9662488)	1
2	cafe	None	Country Style	POINT (-76.4835513 44.2569218)	1
3	cafe	coffee_shop	Tim Hortons	POINT (-76.5066058 44.2598381)	1
4	cafe	coffee_shop	Starbucks	POINT (-76.9346255 40.2387067)	1

In [19]:

us_geo = geopandas.read_file('us_20m.json')
us_geo.crs = {'init' :'epsg:4326'}

In [20]:

us_geo = us_geo[(us_geo.NAME != "Hawaii") & (us_geo.NAME != "Alaska") & (us_geo.NAME != "Puerto Rico")]
us_geo.plot()
plt.show()

In [21]:

cafes = geopandas.GeoDataFrame(cafes)
cafes.crs = {'init' :'epsg:4326'}
cafes = cafes.to_crs({'init': 'epsg:3395'})
us_geo = us_geo.to_crs({'init': 'epsg:3395'})

In [22]:

base = us_geo.plot(color='#e9eeee', edgecolor='#555555', figsize=(20,20))
cafes.plot(ax=base, marker='o', color='green', markersize=1, alpha=0.5)
plt.show()

In [23]:

base = us_geo.plot(color='#e9eeee', edgecolor='#555555', figsize=(20,20))
cafes[cafes['cuisine']!='donut'].plot(ax=base, marker='o', color='green', markersize=1, alpha=0.5)
cafes[cafes['cuisine']=='donut'].plot(ax=base, marker='o', color='red', markersize=5, alpha=1)
plt.show()

Finding the cafe's nearest city or town¶

In [24]:

cafes = cafes.to_crs({'init': 'epsg:4326'})
us_geo = us_geo.to_crs({'init': 'epsg:4326'})

In [25]:

from shapely.ops import nearest_points
multipoint =  cities.geometry.unary_union
def near(point, pts=multipoint):
    nearest = cities.geometry == nearest_points(point, pts)[1]
    return cities[nearest].City.get_values()[0]

In [26]:

cafes_sample = cafes.sample(50)
cafes_sample['nearest'] = cafes_sample.apply(lambda cafe: near(cafe.geometry), axis=1)
cafes_sample[['name', 'nearest', 'geometry']]

Out[26]:

	name	nearest	geometry
19188	Dooney's Cafe	niagara falls	POINT (-79.42479849999999 43.6623165999999)
15660	Firelands Café	sandusky	POINT (-82.59462509999999 41.39778199999985)
5068	Starbucks	fort collins	POINT (-105.0921089 40.55302779999984)
9178	Mitchell's Coffee House	lakeland	POINT (-81.9550215 28.04630939999938)
7915	Bagel and More	new york	POINT (-74.00000740000002 40.73848729999983)
4036	Starbucks Coffee	oakland	POINT (-122.273898 37.80044439999977)
12121	Starbucks	mount vernon	POINT (-73.8283531 40.95044309999985)
14026	The Wydown	washington	POINT (-77.03219110000001 38.91645309999979)
7614	Belle Epicurean	seattle	POINT (-122.2920724 47.62608069998335)
17372	Cafe Ross	san anselmo	POINT (-122.5555492 37.96018919999976)
15504	None	golden	POINT (-105.2218426 39.75571879999981)
4858	Lazy Daisy's	niagara falls	POINT (-79.31956969999999 43.67267709999989)
13221	Heirloom Salad Company	iowa city	POINT (-91.53248619999998 41.65992709999987)
5497	Starbucks	morgan hill	POINT (-121.6436148 37.11249939999973)
4587	Coffee Time	erie	POINT (-80.54027480000002 43.4857523999999)
10416	aroma espresso bar	niagara falls	POINT (-79.3938836 43.64589749999989)
1439	Starbucks Coffee	niagara falls	POINT (-79.38845619999999 43.6701574999999)
11772	Tim Hortons	niagara falls	POINT (-79.18708129999999 43.7833634999999)
5148	Sweet Sue's Bakeshop	parole	POINT (-76.5314048 38.70842029999979)
20659	Cacao Coffee House	santa monica	POINT (-118.4544201 34.04468719999961)
19773	Subway	morgan hill	POINT (-121.639992 37.11093679999973)
1723	The Bean	las cruces	POINT (-106.7955689 32.27744659999957)
4756	Tim Hortons	niagara falls	POINT (-79.6083119 43.5715151999999)
7973	Blue Dot Cafe & Coffee Bar	alameda	POINT (-122.2543493 37.76682479999977)
4452	Tim Hortons	ogdensburg	POINT (-75.8779526 45.30398529999992)
1682	Starbucks	simpsonville	POINT (-82.25589590000001 34.71154639999966)
12911	Starbucks	jacksonville beach	POINT (-81.52557749999998 30.25847729999948)
20467	Ramen Yamadaya	beverly hills	POINT (-118.4430765 34.05781379999963)
15700	Starbucks	newberg	POINT (-123.17308 45.22743829999992)
11611	Bubble Tea	university city	POINT (-90.307902 38.65640029999979)
3876	Fraticelli's Authentic Italian Grill	niagara falls	POINT (-79.38241399999998 43.8638979999999)
9824	Starbucks	guttenberg	POINT (-73.9546503 40.76999359999984)
13241	Starbucks	south salt lake	POINT (-111.9002395 40.72960419999983)
13378	Biggby Coffee	okemos	POINT (-84.4209608 42.71937419999987)
1602	The Coffee House	gallup	POINT (-108.7425899 35.52710149999967)
6107	North Star Coffeehouse	portland	POINT (-122.681791 45.57780759999992)
17066	Sweet Phee's	astoria	POINT (-124.0547878 46.34944119999994)
6657	Iris Cafe	newport	POINT (-84.51223830000001 39.1106949999998)
10078	Over Easy Cafe	asheville	POINT (-82.55205099999999 35.59628989999968)
6013	Starbucks	la vista	POINT (-96.09052419999999 41.26213729999984)
1169	Meshuggah Cafe	university city	POINT (-90.30181019999999 38.65578739999978)
8122	Oak Tree Teriyaki	shoreline	POINT (-122.3439809 47.70209019998354)
4541	Zaya Cafe	oakland	POINT (-122.2689682 37.80750479999976)
12264	Market Square Bakehouse	amesbury	POINT (-70.93043590000001 42.85812609999989)
2091	Bean Traders	durham	POINT (-78.93879529999998 35.90966479999969)
3367	Kaneko Dining	salem	POINT (-123.0271285 44.93283059999991)
18961	Starbucks	wellesley	POINT (-71.2916949 42.30060659999987)
20441	Starbucks	ferndale	POINT (-123.1211255 49.17761419998703)
2100	Zoo Bar Cafe	washington	POINT (-77.05554120000001 38.9298749999998)
8457	Espresso	anacortes	POINT (-122.6124295 48.4966675999855)

In [ ]:

Problems¶

How would you compute the distance from a given cafe to the nearest Starbucks?

References¶

Geopandas mapping: http://geopandas.org/mapping.html