PyData TT Banner Image

makepath presented at the 10th meetup of PyData Trinidad and Tobago (PyData T&T).

PyData Trinidad and Tobago

PyData T&T is a welcoming community for locals interested in data science, analytics, machine learning, or deep learning. It provides opportunities for technology industry professionals and enthusiasts of all skill levels to learn about the latest developments in the Open Source Python and data science community.

The PyData scene in Trinidad and Tobago is ripe for explosion on a global scale. This is largely due to the rising talent in the region and a focus on balancing both first and third world needs. 

With continued encouragement, support, and access to technical resources and mentors, the Python scene in Trinidad and Tobago will take off in an exciting way.

For the 10th Meetup of PyData T&T, makepath Co-Founder and Principal, Brendan Collins, demonstrated how the open source libraries xarray-spatial and RTXpy can be used for spatial data science. 

Open Source GIS and Python for Geospatial

Using a Geographic Information System (GIS) is somewhat analogous to performing spatial data science using Python. 

The xarray-spatial library, which Brendan created, provides tools for solving common geospatial problems as part of the broad ecosystem of Open Source GIS tools. 

A GIS can be used independently or in tandem with open source GIS libraries like xarray-spatial. 

Common geospatial problems can include:

  • Calculating distances between locations (e.g. distance between electric vehicle charging stations along frequently traveled interstate routes)
  • Evaluating land cover change over time (e.g. rainforest to bare soil)
  • Determining the most profitable location to for a new store (e.g. high traffic versus up-and-coming areas)

xarray-spatial benefits from leveraging additional Open Source libraries including:

  • Numba: To speed up numeric calculations.
  • Dask: To scale calculations across different machines.

RTXpy provides xarray-spatial its own c extension for ray tracing using CUDA.

This allows xarray-spatial to quickly address complex spatial data science questions with its raster analysis functions. 

Getting Started with Xarray-Spatial

The best way to get started using xarray-spatial is to review the User Guide.

The xarray-spatial User Guide includes notebooks of examples and test data to help users learn how to leverage the library and key raster analysis tools, such as the multispectral or pathfinding tools. 

You can also find xarray-spatial examples as part of Microsoft’s Planetary Computer GitHub tutorials.

Xarray-Spatial Planetary Computer Notebooks

Even if you don’t know much about Open Source GIS, xarray-spatial can help you perform the same common raster analysis that a GIS can.

Want to Learn More?

Want to learn more about how xarray-spatial works for spatial data science? You can watch the entire session on the PyData TT Youtube Channel

Want to start contributing to xarray-spatial? Start with makepath’s guide on how to get started with Open Source

Have any questions, ideas or thoughts to share? Let’s connect at