Clarification on "conditional = None" parameter and on Cebra Multianimal with no labels

[SofiaTorchia]

Hi,

I am currently working with some 1-p data on multiple mice for which I also have a one-dimensional categorical behavioral label over time (e.g.: ['flight','flight','fight',None,None,..,None, 'risk-assessment','risk-assessment'..]).
I would like to use Cebra-Behavior Multianimal where no information about time is used to model the neural embedding (supervised contrastive learning where the auxiliary variable would exclusively be the categorical behavioral label). Should I set conditional = None as parameter for Cebra?
Is conditional='delta' used for Cebra-Behavior with only continuous behavioral data as auxiliary variable?

Another point:
I noticed that If I don't input any sort of behavioral label to Cebra Multianimal, I get the following error:
RuntimeError: No label: labels are needed for alignment in the multisession implementation..
Does this mean it's not possible to run Cebra-Time Multianimal as you would do with Cebra-Time Single Session? Or is it enough to run Cebra-Behavior Multianimal and input the same label for each animal (e.g.: [0,0,0,0,...0] for all animals) to overcome this issue?

Thank you so much for being active on this discussion platform and provide support to Cebra users!

Sofia