Why yet another BCI framework?
We needed something that:
fits well within the Python datascience and machine learning ecosystem
has a permissive license, allowing commercial applications
works both offline and online
enables quick prototyping
is not only about EEG data or even about the brain, and that can works with any time series
We were not able to find a framework that satisfied all of our requirements, so we built it.
Timeflux is primarily developed by Pierre Clisson and Raphaëlle Bertrand-Lalo, with the support of awesome contributors.
The introduction paper has been be published in the Proceedings of the 2019 Graz BCI Conference.