search

LabeliaLabs / distributed-learning-contributivity

Simulate collaborative ML scenarios, experiment multi-partner learning approaches and measure respective contributions of different datasets to model performance.
https://www.labelia.org
Apache License 2.0
56 stars 12 forks source link

Unable to run single partner learning approach #351

Closed HeytemBou closed 2 years ago

HeytemBou commented 3 years ago

https://github.com/SubstraFoundation/distributed-learning-contributivity/blob/4fb247a35ae40ce27790b888cc6e4cb52134fa68/mplc/multi_partner_learning/basic_mpl.py#L350

@arthurPignet Is this comment still valid or is it no longer possible to use the Single Partner Learning approach when running a scenario?

arthurPignet commented 3 years ago

I just checked, the mpl method single partner learning does exist, (in basic mpl) but it is not reachable. I will fix it. However this comment is not really relevant anymore, the method raises an error if only one partner is provided, and doesn't fall back to the dedicated single partner class