Amsterdam Transport

import osmnx as ox
import networkx as nx
import pickle
import matplotlib.pyplot as plt
import pandas as pd

# 4.0 plotting the route
ox.config(use_cache=True, log_console=True) 
city = ox.graph_from_place('Amsterdam', retain_all=False, truncate_by_edge=False, 
                        simplify=True, custom_filter='["waterway"~"canal"]') 

# the starting and finishing point is AMS institute and the national museum
location = ox.geocode("Kattenburgerstraat 5, Amsterdam, Netherlands") 
location_to = ox.geocode("Museumstraat 1, Amsterdam, Netherlands") 
nodes = ox.distance.nearest_nodes(city, location[1], location[0], return_dist=True) 
nodes_to = ox.distance.nearest_nodes(city, location_to[1], location_to[0], return_dist=True)

# finding the route along the canal
path = ox.shortest_path(city, nodes[0], nodes_to[0]) 
# print(path)

# plotting
pt = ox.graph_to_gdfs(city, edges=False).unary_union.centroid 
bbox = ox.utils_geo.bbox_from_point(location, dist=5000) 
fig, ax = ox.plot_graph_route(city, path, bbox=bbox, show=False, close=False) 

# calculating the cordinates of the center point
lat = 0 
lon = 0 
for points in path: 
    lat += city.nodes[points]['x'] 
    lon += city.nodes[points]['y'] 
lat_center = lat / len(path) 
lon_center = lon / len(path) 
center_point = (lon_center, lat_center) 
print(center_point)

# plotting
ax.scatter(lon_center, lat_center, c='blue') 
plt.show()

# finding the whole map and finding the center node from the map
city_whole = (ox.graph_from_place('Amsterdam, Netherlands', simplify=True)) 
nodes_centre = ox.distance.nearest_nodes(city_whole, center_point[1], center_point[0], return_dist=True)

# plotting
pt = ox.graph_to_gdfs(city, edges=False).unary_union.centroid 
bbox = ox.utils_geo.bbox_from_point(center_point, dist=5000) 
fig, ax1 = ox.plot_graph(city_whole, bbox=bbox, show=True, close= False) 
ax1.scatter(center_point[0], center_point[1], c='blue') 
plt.show()

#finding the bus and tram stops near the starting point
try: 
    bus_stop = ox.features.features_from_point(location, tags={'public_transport': 'stop_position'}, dist=500) 
    bus_stop.to_csv("bus_stop.csv") 
    print(bus_stop.name) 
except: 
    print("no station in @‘dist’ metres")

# finding the bus and tram stops near the ending point
try: 
    bus_stop_to = ox.features.features_from_point(location_to, tags = {'public_transport': 'stop_position'}, dist=500) 
    bus_stop_to.to_csv("bus_stop_to.csv") 
    print(bus_stop_to.name) 
except: 
    print("no station in @‘dist’ metres")

# finding the bus route,but the code is not showing any route, reason unknown
tags = {'route': 'bus'}
gdfox = ox.features_from_point(location_to, tags, dist=1000)
print(gdfox)

# finding the map
graph = ox.graph_from_point(center_point, dist=5000)

# finding the nodes close to the start/center/end point
nodes = ox.distance.nearest_nodes(graph, location[1], location[0], return_dist=True)
nodes_to = ox.distance.nearest_nodes(graph, location_to[1], location_to[0], return_dist=True)
nodes_cen = ox.distance.nearest_nodes(graph, center_point[1], center_point[0], return_dist=True)

# calculate centrality
start_centrality = nx.closeness_centrality(graph, nodes[0])
cen_centrality = nx.closeness_centrality(graph, nodes_cen[0])
end_centrality = nx.closeness_centrality(graph, nodes_to[0])
print(start_centrality,cen_centrality,end_centrality)

# finding the map near the ending point
plt.close('all') 
graph = ox.graph_from_point(location_to, dist=2100)

# extracting all the cafes and restaurants within 750 radius
poi = ox.features.features_from_point(location_to, tags={'amenity': ['cafe','restaurant']}, dist=750) 
target_node = ox.distance.nearest_nodes(graph, location_to[1], location_to[0], return_dist=True) 
pt = ox.graph_to_gdfs(city_whole, edges=False).unary_union.centroid 
bbox = ox.utils_geo.bbox_from_point(location_to, dist=2100) 
fig2, ax2 = ox.plot_graph(city_whole, bbox=bbox, show=False, close=False)

# examine the shortest distance, if it is within 10 minutes walk(750m)
name = []
i = 0
for row in poi.T.T.itertuples():
    try:
        orig = getattr(row, 'geometry')
        name1 = getattr(row, 'name')
        orig_node = ox.distance.nearest_nodes(graph, orig.x, orig.y, return_dist=True)
        length = nx.shortest_path_length(G=graph, source=orig_node[0], target=target_node[0], weight='length')
        if length <750:
            name.append(name1)
            ax2.scatter(orig.x, orig.y, c='blue')
            i += 1
            if i == 1:
                listing1 = pd.DataFrame(
                    {"amenity":[getattr(row, 'amenity')],"length":[length],"name":[getattr(row, 'name')],
                     "geometry":[getattr(row, 'geometry')],"description":[getattr(row, 'description')]})
            else:
                listing = pd.DataFrame(
                    {"amenity": [getattr(row, 'amenity')], "length": [length], "name": [getattr(row, 'name')],
                     "geometry": [getattr(row, 'geometry')], "description": [getattr(row, 'description')]})
                listing1 = pd.concat([listing1, listing], ignore_index=True)

    except:
        pass
listing1.to_csv("listing.csv")

# plotting
plt.show() 
plt.savefig("4.jpg") 
print(name)