search

BouchardLab / nsds_lab_to_nwb

Python package to convert NSDS Lab data to NWB files.
https://nsds-lab-to-nwb.readthedocs.io/en/latest/
0 stars 4 forks source link

Fix filename convention & organization for block metadata files #3

Closed jihyunbak closed 3 years ago

jihyunbak commented 3 years ago

Perhaps organize the block metadata files by rats?

For example,

All changes in the metadata repo should be reflected in MetadataManager.

What else?

jthermiz commented 3 years ago

I like the alternative organization R00/R00_B00.yaml

Do we want to continue to use R00_B00 convention? What happens when we reach 99+ 1? Do we increment to R100 or go back to R00?

KrisBouchard commented 3 years ago

Lets go with R####_B### just to be safe.

If we get to R10000, I will simply retire.

Make sense?

-- KB (he/him/his)

Kristofer Bouchard, Ph.D. Computational Biology Staff Scientist

PI, Neural Systems and Data Science Lab http://bouchardlab.lbl.gov, Biological Systems & Engineering Division, LBNL Acting Lead, Computational Biosciences Group https://crd.lbl.gov/computational-biosciences, Computational Research Divisions, LBNL

Assistant Adjunct Professor Helen Wills Neuroscience Institute & Redwood Center for Theoretical Neuroscience, UC Berkeley

Kavli Institute for Fundamental Neuroscience UC San Francisco

On Fri, Dec 4, 2020 at 3:39 PM John Hermiz notifications@github.com wrote:

I'll like the alternative organization R00/R00_B00.yaml

Do we want to continue to use R00_B00 convention? What happens when we reach 99+ 1? Do we increment to R100 or go back to R00?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/BouchardLab/nsds_lab_to_nwb/issues/3#issuecomment-739078354, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABZFA77HFQE3W7AP23SOXADSTFXLLANCNFSM4UM5W5QA .

jihyunbak commented 3 years ago

Update:

Note on the animal/block name conventions: essentially, these conventions should be implemented on the metadata side (and when saving the raw data files). The builder simply takes two strings, like ("R00100", "B001"), and it won't care how the strings are formatted inside. So we can use (animal, block) names like ("MyLittleRat", "TestBlock") without any problem, as long as there are matching raw data files. This flexibility will make thing easier for simulated datasets, because you can just treat it like an experiment where the animal name is "Simulation".