LossFunction

LearningHorse.LossFunction.mseFunction
mse(y, t)

Return the mean square error:mean((y .- t) .^ 2)

Example

julia> loss = mse
mse (generic function with 1 method)

julia> y, t = [1, 3, 2, 5], [1, 2, 3, 4]
([1, 3, 2, 5], [1, 2, 3, 4])

julia> loss(y, t)
0.75
source
LearningHorse.LossFunction.ceeFunction
cee(y, t, dims = 1)

Return the cross entropy error:mean(-sum(t.*log(y)))

Example

julia> loss = cee
cee (generic function with 1 method)

julia> y, t = [1, 3, 2, 5], [1, 2, 3, 4]
([1, 3, 2, 5], [1, 2, 3, 4])

julia> loss(y, t)
-10.714417768752456
source