the classification criterion doesn't factor in the uncertainty - does this mean ignore the uncertainty for classification?

[sgbaird]

i.e. log_std is an unused parameter in the classification criterion:

CrabNet/utils/utils.py

Lines 263 to 265 in a5be06f

	def BCEWithLogitsLoss(output, log_std, target):
	loss = nn.functional.binary_cross_entropy_with_logits(output, target)
	return loss

If this is the case, should the uncertainty output from CrabNet be ignored by the user during classification? In other words, are the uncertainty values essentially just a bunch of random numbers for classification?