"Here, the graph files gt.p and prop.p are all in the same format as what we used in Sat2Graph - a python dictionary where each key is the coordinate of a vertex (denoted by x) and the corresponding value is a list of x's neighboring vertices."
All rights for the topo algorithm and its documentations go to the authors and creators of Sat2Graph (Songtao He and his collaborators). Link to the main repo.
txt2RoadGraph is a tool I have created for conversions from text-based graphs to a pickle file containing a RoadGraph object, suitable for this TOPO evaluation. Simply put your vertices.txt and edges.txt in the same directory and convert them to a pickle file:
python3 txt2RoadGraph.pyThen you are all set to start the evaluation:
python3 main.py -graph_gt data/chicago_osm -graph_prop data/chicago_james -output toporesult.txt| Parameters | Note |
|---|---|
| Propagation Distance | 300 meters for large tiles and 150 meters for small tiles (see line 127-130 in main.py) |
| Propagation Interval | Default is 5 meters. Config with -interval flag. |
| Matching Distance Threshold | Default is 10 meters. Config with -matching_threshold flag. |
| Matching Angle Threshold | 45 degrees |
| One-to-One Matching | True |
- hopcroftkarp
- rtree
- pyproj (for txt2RoadGraph)