[Python] Isochrones with both walk and subway in OSMnx

I live in Paris (France) and I would like to compute isochrones from a given point in the city.

I would like to compute these isochrones by both walking and taking subways solutions (not one or the other way, but the two simultaneously. Something that we could get by using Google Maps)

The OSMnx package seems to be able to compute isochrones.

I see in the documentation :

• 
  a solution to compute isochrone by walking only (through the definition of a custom meters_per_minute variable)
  
  
• 
  a solution to import New York's subway graph (see the section called "Get other networked infrastructure types")
  
  Basically, the New York subway graph can be obtained this way:
  
  # get NY subway rail network
  G = ox.graph.graph_from_place(
  "New York, New York, USA",
  retain_all=False,
  truncate_by_edge=True,
  simplify=True,
  custom_filter='["railway"~"subway"]',
  )
  
  fig, ax = ox.plot.plot_graph(G, node_size=0, edge_color="w", edge_linewidth=0.2)

But I fail to see if it's possible to compute an isochrone that use both walking and subway solution.

Since I do not see on OpenStreetMap an option to compute a distance "by subway", perhaps my problem cannot be solved by using an API on OpenStreetMap.

Have you ever tried to solve this kind of problem?

Continue reading...