This program generates potential sidewalk network from OpenStreetMap network data.

• OSM2Geojson: http://tyrasd.github.io/osmtogeojson/
• LatLng visualization: http://www.darrinward.com/lat-long/?id=592078
• Geojson visualization: http://geojson.io/
• Geojson visualization: https://www.mapbox.com/editor/#data

This project uses the following python libraries as well as the C/C++ libraries that they depend on.

• Basemap >= 1.0.7
• Shapely >= 1.5.8
• Rtree >= 0.8.2

I think, especially on Windows, the easiest way to go is to use Anaconda Python (https://store.continuum.io/cshop/anaconda/), a free Python distribution from Continuum.

To install requirements, just use the conda command (e.g., conda install basemap). Note, however, at the time of writing this README, GEOS (a C++ library for GIS stuff) cannot be installed on a Windows machine via conda. So instead, I install it through OSGeo4W bundle (http://trac.osgeo.org/osgeo4w/).

Another hiccup I encountered with installing dependencies is installing Shapely and Rtree. For Shapely, I followed the instruction here to install it: http://gis.stackexchange.com/questions/38899/geos-and-shapely-installation-on-windows . Similarly, I installed Rtree using the whl in the following link: http://www.lfd.uci.edu/~gohlke/pythonlibs/#rtree

Example:

1. Uninstall Shapely (pip uninstall shapely)
2. Download the Shapely wheel file from the following link: http://www.lfd.uci.edu/~gohlke/pythonlibs/#shapely
3. Run pip install wheel
4. Run pip install THE_WHEEL_FILENAME

• Zachary Lawrence (https://github.com/zacharylawrence)
• Anthony Li (https://github.com/tongning)
• Kotaro Hara