Cityseer: Urban Network Analysis

4.24.1 · active · verified Thu Apr 16

Cityseer is a Python library providing computational tools for network-based pedestrian-scale urban analysis. It enables users to model urban environments, analyze accessibility, connectivity, and other urban metrics using spatial data and graph theory. The current version is 4.24.1, and it typically sees regular updates, often with significant changes between major versions.

Common errors

```
ModuleNotFoundError: No module named 'cityseer.graph'
```
cause Attempting to import or use the `Graph` object or its related modules from older (v3) Cityseer versions in v4+.

fix The `Graph` object has been removed in v4. All core network operations and data are now managed by the `cityseer.City` object. Use `City.from_osmnx(graph)` or similar constructors to create a `City` object, and then access network data and methods through it (e.g., `city.nodes`, `city.edges`).
```
AttributeError: 'City' object has no attribute 'graph_from_bbox'
```
cause Calling a v3 standalone function (like `graph_from_bbox`) as a method on the new `City` object in v4+.

fix In v4+, `graph_from_bbox` and similar functions have been absorbed into the `City` constructor. Instead, initialize the `City` object directly with the bounding box or use `City.from_osmnx` with an already created OSMnx graph. E.g., `city = City('MyCity', bounding_box=(miny, minx, maxy, maxx), crs='EPSG:4326')`.
```
pyproj.exceptions.CRSError: Invalid CRS: ...
```
cause The provided CRS string or object is not recognized or is malformed, often due to typos or incorrect EPSG codes.

fix Double-check the CRS string for correctness. Common valid options are `EPSG:4326` (WGS84 lat/lon) or local projected CRS codes (e.g., `EPSG:27700` for OSGB36). Ensure `pyproj` is up-to-date. If loading from a GeoDataFrame, use `gdf.crs` directly.

Warnings

breaking Version 4.0.0 introduced significant breaking changes, refactoring the library around a central `City` object. Many standalone functions and older `Graph` objects were removed or moved into methods of the `City` class.
Fix: Review the official v4 migration guide. Replace direct `graph` object manipulation and standalone functions (e.g., `graph_from_bbox`, `accessibility`) with methods on the `cityseer.City` instance (e.g., `City.from_osmnx`, `city.analyze_accessibility`).
gotcha Coordinate Reference System (CRS) handling is strict. You must explicitly provide a valid CRS (e.g., `crs='EPSG:4326'` for WGS84 or a projected CRS) when initializing a `City` object or importing data.
Fix: Always pass the `crs` argument with a valid EPSG code or WKT string to `City` constructors (e.g., `City('place', crs='EPSG:4326')` or `City.from_osmnx(G, crs=G.graph['crs'])`). Ensure your input data also has a defined CRS.
gotcha Analyzing very large urban areas or extremely dense networks can lead to high memory consumption and long processing times. Cityseer is designed for pedestrian-scale analysis.
Fix: For initial development and testing, work with smaller, representative areas. Consider simplifying the network or processing in smaller chunks for very large regions. Monitor memory usage and execution time.

Install

pip install cityseer Install latest version

Imports

City
wrong:
```
from cityseer.city import City
```
correct:
```
from cityseer import City
```
The City class is directly available under the top-level package namespace in v4+.

Quickstart

This quickstart demonstrates how to create a `Cityseer` `City` object from an `OSMnx` graph and perform a basic accessibility analysis. It highlights the primary entry point (`City.from_osmnx`) and a common analysis method (`analyze_accessibility`). Ensure `osmnx` is also installed (`pip install osmnx`).

import osmnx as ox
from cityseer import City

# Define a place and retrieve its street network using OSMnx
place_name = 'Piedmont, California, USA'
G = ox.graph_from_place(place_name, network_type='walk')

# Create a Cityseer City object from the OSMnx graph
city = City.from_osmnx(G, crs='EPSG:4326')

# Perform a simple accessibility analysis
accessibility_result = city.analyze_accessibility(target_points=city.nodes.geometry)

# Print a sample of results (e.g., average accessibility)
print(f"Average accessibility (example): {accessibility_result.mean():.2f}")

# To save or visualize, you would typically use city.to_gdfs() or city.draw_network()
# For example, to get nodes with accessibility data:
# nodes_gdf = city.nodes.copy()
# nodes_gdf['accessibility'] = accessibility_result
# print(nodes_gdf.head())

view raw JSON →