Support for lecun normal weight initialization

src/utils.jl

@@ -252,6 +252,48 @@ truncated_normal(rng::AbstractRNG=default_rng(); init_kwargs...) = (dims...; kwa
 
 ChainRulesCore.@non_differentiable truncated_normal(::Any...)
 
+"""
+    lecun_normal([rng], size...) -> Array
+    lecun_normal([rng]; kw...) -> Function
+
+Return an `Array{Float32}` of the given `size` containing random numbers drawn from a truncated normal
+distribution centered on 0 with stddev `sqrt(1 / fan_in)`, where `fan_in` is the number of input units 
+in the weight tensor.
+
+# Examples
+```jldoctest; setup = :(using Random; Random.seed!(0))
+julia> using Statistics
+
+julia> round(std(Flux.lecun_normal(10, 1000)), digits=3)
+0.032f0
+
+julia> round(std(Flux.lecun_normal(1000, 10)), digits=3)
+0.317f0
+
+julia> round(std(Flux.lecun_normal(1000, 1000)), digits=3)
+0.032f0
+
+julia> Dense(10 => 1000, selu; init = Flux.lecun_normal())
+Dense(10 => 1000, selu)  # 11_000 parameters
+
+julia> round(std(ans.weight), sigdigits=3)
+0.319f0
+```
+
+# References
+
+[1] Lecun, Yann, et al. "Efficient backprop." Neural networks: Tricks of the trade. Springer, Berlin, Heidelberg, 2012. 9-48.
+"""
+function lecun_normal(rng::AbstractRNG, dims::Integer...; gain::Real=1)
+  std = Float32(gain)*sqrt(1.0f0 / first(nfan(dims...))) # calculates the standard deviation based on the `fan_in` value
+  return truncated_normal(rng, dims...; mean=0, std=std)
+end
+
+lecun_normal(dims::Integer...; kwargs...) = lecun_normal(default_rng(), dims...; kwargs...)
+lecun_normal(rng::AbstractRNG=default_rng(); init_kwargs...) = (dims...; kwargs...) -> lecun_normal(rng, dims...; init_kwargs..., kwargs...)
+
+ChainRulesCore.@non_differentiable lecun_normal(::Any...)
+
 """
     orthogonal([rng], size...; gain = 1) -> Array
     orthogonal([rng]; kw...) -> Function