MovingPandas v0.9 released!

The latest v0.9 release is now available from conda-forge.

This release contains some really cool new algorithms:

Trajectory smoothing (Kalman filter)
Spatiotemporal trajectory generalization (Top-Down Time Ratio)
Trajectory cleaning (statistical outliers in numerical columns)

The Kalman filter in action on the Geolife sample: smoother, less jiggly trajectories.

Top-Down Time Ratio generalization aka trajectory compression in action: reduces the number of positions along the trajectory without altering the spatiotemporal properties, such as speed, too much.

These new algorithms were contributed by Lyudmil Vladimirov and George S. Theodoropoulos.

Behind the scenes, Ray Bell took care of moving testing from Travis to Github Actions, and together we worked through the steps to ensure that the source code is now properly linted using flake8 and black.

Being able to work with so many awesome contributors has made this release really special for me. It’s great to see the project attracting more developer interest.

As always, all tutorials are available from the movingpandas-examples repository and on MyBinder: