Update the math formula for LogLoss to match the parameters.

PiperOrigin-RevId: 227875597
This commit is contained in:
Scott Zhu 2019-01-04 10:24:02 -08:00 committed by TensorFlower Gardener
parent daa75aff18
commit d183818d4d

View File

@ -503,7 +503,7 @@ class CategoricalHinge(Loss):
class LogLoss(Loss):
"""Computes the log loss between `y_true` and `y_pred`.
logloss = -y log(p) - (1-y) log(1-p)
logloss = - y_true * log(y_pred) - (1 - y_true) * log(1 - y_pred)
Usage: