Binary cross-entropy loss

From GISAXS
Revision as of 16:37, 3 February 2023 by KevinYager (talk | contribs)
Jump to: navigation, search

The binary cross-entropy loss is given by:

for training examples (indexed by ) where is the class label (0 or 1) and is the prediction for that example (i.e. the predicted probability that it is a positive example). Thus is the probability that it is a negative example. Note that we are adding