Why do we expand labels?

[conorturner]

MPL-pytorch/data.py

Line 138 in 7fb5b40

if args.expand_labels or args.num_labeled < args.batch_size:

Hey all,

I've been working my way through this implementation and cannot work out why the expand labels options exists? It seems that even if the labels aren't expanded the data loader will loop anyway.

Can anyone explain why this is needed?