Gelu
nnf_gelu(input, approximate = "none")
input
: (N,*) tensor, where * means, any number of additional dimensionsapproximate
: By default it's none, and applies element-wise x*pnorm(x), if 'tanh', then GELU is estimated. See GELU for more info.Applies element-wise the function
where is the Cumulative Distribution Function for Gaussian Distribution.
Useful links