bayesian-neural-network-blogpost icon indicating copy to clipboard operation
bayesian-neural-network-blogpost copied to clipboard

Published 20 hours ago • verified publisher icon kyle-dorman

about the variance of classification task

Open ShellingFord221 opened this issue 5 years ago • 0 comments
trafficstars

Hi, when modeling aleatoric uncertainty, why is pred_var, predicted logit values and variance's shape: (N, C + 1)? Shouldn't it be (N, C * 2)?

ShellingFord221 avatar Dec 23 '19 14:12 ShellingFord221