You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am finding when generating multiple splits for BioSoundSegBench that sometimes vak.prep.split.split.train_test_dur_split_inds returns indices for a training set that is just slightly smaller than the target duration, e.g. 599.5 seconds instead of 600s.
Not sure how this happens -- we should have a post-condition to guard against it -- but generating multiple splits with a fixed dataset seems like a good start for replicating it
The text was updated successfully, but these errors were encountered:
I am finding when generating multiple splits for BioSoundSegBench that sometimes
vak.prep.split.split.train_test_dur_split_inds
returns indices for a training set that is just slightly smaller than the target duration, e.g. 599.5 seconds instead of 600s.Not sure how this happens -- we should have a post-condition to guard against it -- but generating multiple splits with a fixed dataset seems like a good start for replicating it
The text was updated successfully, but these errors were encountered: