Epsilon change in normalise for stability (#2421)

* epsilon change for stability

* Change comment for eps

Co-authored-by: Carlo Lucibello <carlo.lucibello@gmail.com>

---------

Co-authored-by: Carlo Lucibello <carlo.lucibello@gmail.com>
Co-authored-by: Carlo Lucibello <carlo.lucibello@unibocconi.it>

Author: billera

src/layers/stateless.jl

@@ -1,10 +1,10 @@
 
 """
-    normalise(x; dims=ndims(x), eps=1e-5)
+    normalise(x; dims=ndims(x), eps=1f-5)
 
 Normalise `x` to mean 0 and standard deviation 1 across the dimension(s) given by `dims`.
 Per default, `dims` is the last dimension. 
-`eps` is a small term added to the denominator for numerical stability.
+`eps` is a small term added to the variance for numerical stability.
 
 # Examples
 ```jldoctest
@@ -34,10 +34,11 @@ julia> isapprox(std(y; dims=1, corrected=false), ones(1, 10), atol=1e-5)
 true
 ```
 """
-@inline function normalise(x::AbstractArray; dims=ndims(x), eps=ofeltype(x, 1e-5))
+@inline function normalise(x::AbstractArray; dims=ndims(x), eps=1f-5)
   μ = mean(x, dims=dims)
-  σ = std(x, dims=dims, mean=μ, corrected=false)
-  return @. (x - μ) / (σ + eps)
+  σ² = var(x, dims=dims, mean=μ, corrected=false)
+  ε = ofeltype(x, eps)
+  return @. (x - μ) / sqrt(σ² + ε^2)
 end
 
 """