Hi, this GitHub repository contains Python code for using the Dimensionality Transcending (DT) method for merging datasets defined in a SPD manifold. The paper presenting this method is still under review, but if you want to know more about it, feel free to ask me via e-mail (pedro.rodrigues@melix.org) or go directly to Chapter 5 of my Ph.D. thesis (link here)

As you will notice, this repository is still very simple and contains just one script for running an example with motor imagery data. If you need assistance and/or want to see more examples, please feel free to contact me or raise an issue. I'm planning to include more stuff here very soon.

To run the code available, you will have to install a few extra packages. The most important ones being:

​ -- MOABB, available here (you should probably use my fork to make things work)

​ -- Riemann-lab, available here

If you would like to know more about the Riemannian Procrustes Analysis (RPA) procedure that is used in DT, you're in luck. We have a GitHub repo with explanations and code (link here).