speed-up with modin ?

may we can consider using modin[1] for speed up..

[1] https://modin.readthedocs.io/en/latest/

Looks interesting. We should definitely try that and if it works as advertised include it (maybe as an optional requirement?).

added needs discussion label

added 1 deleted label

added 1 deleted label

removed 1 deleted label

Just tried to throw modin on dios, as this would currently be our man target for such an optimization, and it turned out that the package is not as compatible to pandas as I hoped. Haven't checked thoroughly but the first error was raised by a call to df.at.

added 1 deleted label

Closing, as we lost the ability to work with modin, as we introduced custom data structures.

closed