Subscribe to DSC Newsletter

What Is Your Favorite Python Library For Visualizing Geospatial Data?

We've looked into various geospatial data visualization libraries such as:

  • PySAL
  • OSMnx
  • Folium
  • Mapbox GL
  • GDAL
  • Rasterio
  • Fiona
  • PyProj
  • Descartes

However, since we cater to large enterprises with utility grade data use cases, only GeoPandas and Shapely meet the following criteria:

  1. easy to set-up
  2. stable release (no experimental or early releases)
  3. easily available in both pip and Anaconda
  4. sufficient amount of documentation available
  5. a big community behind it and many users
  6. tests/runs successfully on both IPython command-line and Jupyter Notebook

GeoPandas

http://geopandas.org/

Shapely

https://github.com/Toblerity/Shapely
https://shapely.readthedocs.io/en/latest/

What's your favorite library? And why? Please leave a comment, connect with me via LinkedIn or send an email to [email protected] Thanks!

Views: 1129

Reply to This

Replies to This Discussion

Hi Rafael, what were the pros and cons of the various libraries that you tried? 

The pros of GeoPandas and Shapely are that they fulfill all criteria that matter to corporations. The cons are that some other libraries outshine GeoPandas and Shapely in selected areas. For example, I personally like using Folium, but that library falls through the cracks because it does not meet all our enterprise criteria, and that is:

  1. easy to set-up
  2. stable release (no experimental or early releases)
  3. easily available in both pip and Anaconda
  4. sufficient amount of documentation available
  5. a big community behind it and many users
  6. tests/runs successfully on both IPython command-line and Jupyter Notebook

If you are a singular user, some of those criteria might matter less to you. However, if you work in a large and/or highly regulated enterprise, they do all. If you work for a bank, insurance or manufacturer, it's not only a matter of what you like using but what makes sense from a corporate viewpoint.

More specifically, we do not measure by categorical values (library x fulfills criteria 1. -6. YES/NO) but numerically on a scale between 0 and 10. For example, sufficient amount of documentation available is an issue with nearly all geospatial data visualization libraries we looked into. Sometimes it was a matter of choosing between "barely sufficient" and "close to nothing".

RSS

Videos

  • Add Videos
  • View All

Follow Us

© 2018   Data Science Central ®   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service