I’m writing on behalf of the INCF/OCNS Software Working Group. We host regular sessions on various software of interest to the neuroscience community. The sessions are open to all, and we record them where possible and upload them to the INCF Training space.
Given how useful Datalad is, I was wondering of someone from the Datalad team would please have the time to do a one hour session introducing Datalad and maybe showing a quick demo on how it can be used by the computational modelling community to keep their code, input data, and output data nicely organised and make their projects reproducible?
handbook has various examples and we did give talks in the past tuning them to specific audience, e.g. MILA machine learning folks DataLad on a typical for them dataset. Do you have some representative dataset/code/workflow somewhat specific to “computational modelling community”? may be it could be something which models existing real data in an archive which has data exposed via DataLad , e.g. DataLad Repository of crcns.org or even better DataLad datasets of dandiarchive · GitHub or dandiarchive.org?
I don’t have one myself, unfortunately. I think we do have an example in the NeuroML docs that uses an NWB file to fit a neuronal model, but I’ll have to check if the file lives in a datalad accessible repository—the example currently doesn’t use Datalad to fetch it.
Maybe a general high level introduction to Datalad would be sufficient for a start—just to show folks how they can keep their code and data all nicely managed by Datalad?