[SOLVED] Subdividing streets with interpolate points function

Subdividing streets with interpolate points function

I want to subdivide streets by maximum distance as described in this question and response. My problem is that I do not understand the "Interpolate Points" function output.

My current code is:

# Imports

!pip install osmnx
import osmnx as ox

import networkx as nx

# Enter the place name you want to search
# Find accepted names at https://nominatim.openstreetmap.org/ui/search.html?q=Loop%2C+Chicago%2C+Illinois

place_name = "Macon, Macon County, Illinois, United States"

# Generate a graph of pedestrian circulation in nodes and edges

graph = ox.graph_from_place(place_name, network_type="walk")

fig, ax = ox.plot_graph(graph)

# Interpolate every 10 meters

multiline_r = ox.utils_geo.interpolate_points(graph, 10)

display(multiline_r)

And it gives me a strange output type called "generator" that cannot be transformed into a graph. I have also tried feeding the interpolate points function individual line segments and it gives me this same value. I was expecting some sort of output that I could build into a graph, for example a list of points.

In short, does anybody have a more detailed explanation for implementing the "Interpolate Points" function described in the previous post?

Edit: My solution code is


# imports

!pip install osmnx
import osmnx as ox

import pandas as pd

import geopandas as gpd

!pip install momepy
import momepy

import networkx as nx

import matplotlib.pyplot as plt

# Enter the place name you want to search
# Find accepted names at https://nominatim.openstreetmap.org/ui/search.html?q=Loop%2C+Chicago%2C+Illinois

place_name = "Macon, Macon County, Illinois, United States"

# Generate a graph of pedestrian circulation in nodes and edges

graph = ox.graph_from_place(place_name, network_type="walk")

fig, ax = ox.plot_graph(graph)

# Extract nodes and edges from the graph

nodes, edges = ox.graph_to_gdfs(graph)

# gdf to dataframe

edges_df = pd.DataFrame(edges)

# return geometry column as list of linestrings

geometry_list = edges_df['geometry'].tolist()

display(geometry_list)

# create new evenly spaced edges

lst = [] # initiallize emtpy list

for i in range(len(geometry_list)): # iterate over geometry list
    
    temp_lst = list(ox.utils_geo.interpolate_points(geometry_list[i], 0.0005))
    
    for i in range(len(temp_lst)-1): # append every item in temp list
        temp_edge = [temp_lst[i+1], temp_lst[i]] # create edge
        lst.append(temp_edge)

# Convert coordinate list into linestring list

linestrings = [] # create empty list

for i in range(len(lst)): # iterate over lst
    
    L = ox.utils_geo.LineString(lst[i])
    
    linestrings.append(L)

linestrings

# convert linestrings from list to geodataframe

gdf1 = gpd.GeoDataFrame({'geometry':linestrings},
                                      geometry='geometry', 
                                      crs=3528)

# transform the geopandas file into a networkx graph

G_primal = momepy.gdf_to_nx(gdf1, approach="primal")

f, ax = plt.subplots(1, 2, figsize=(12, 6), sharex=True, sharey=True)
gdf1.plot(color="k", ax=ax[0])
for i, facet in enumerate(ax):
    facet.set_title(("pedestrian", "Graph")[i])
    facet.axis("off")
    #add_basemap(facet) # this wasn't working for some reason so I cut it out
nx.draw(
    G_primal, {n: [n[0], n[1]] for n in list(G_primal.nodes)}, ax=ax[1], node_size=50
)

Solution

I do not understand the "Interpolate Points" function output.

The OSMnx documentation describes the function's output: "points (generator) – tuples of (x, y) floats of the interpolated points’ coordinates"

it gives me a strange output type called "generator"

If you are wondering what a Python generator is: https://stackoverflow.com/a/1756156/7321942

I was expecting some sort of output that I could build into a graph, for example a list of points.

This function returns a generator of points, which you could cast to a list if you'd like.