Layered flux at multiple SDS

[heltonmc]

Appears to be an issue calculating the flux when rho is a vector.

julia> data = Nlayer_cylinder(a = 10.0, l = [1.0, 1.0, 1.0, 2.0], z = 5.0, ρ = [1.0, 2.0])
julia> flux_DA_Nlay_cylinder_CW(data)
ERROR: MethodError: no method matching Float64(::ForwardDiff.Dual{ForwardDiff.Tag{LightPropagation.var"#69#71"{Vector{Float64}, Vector{Float64}, Vector{Float64}, Float64, Vector{Float64}, Vector{Float64}, Float64, Vector{Float64}}, Float64}, Float64, 1})
Closest candidates are:
  (::Type{T})(::Real, ::RoundingMode) where T<:AbstractFloat at rounding.jl:200
  (::Type{T})(::T) where T<:Number at boot.jl:760
  (::Type{T})(::AbstractChar) where T<:Union{AbstractChar, Number} at char.jl:50
  ...
Stacktrace:
 [1] convert(#unused#::Type{Float64}, x::ForwardDiff.Dual{ForwardDiff.Tag{LightPropagation.var"#69#71"{Vector{Float64}, Vector{Float64}, Vector{Float64}, Float64, Vector{Float64}, Vector{Float64}, Float64, Vector{Float64}}, Float64}, Float64, 1})
   @ Base ./number.jl:7
 [2] setindex!(A::Vector{Float64}, x::ForwardDiff.Dual{ForwardDiff.Tag{LightPropagation.var"#69#71"{Vector{Float64}, Vector{Float64}, Vector{Float64}, Float64, Vector{Float64}, Vector{Float64}, Float64, Vector{Float64}}, Float64}, Float64, 1}, i1::Int64)
   @ Base ./array.jl:839
 [3] _kernel_fluence_DA_Nlay_cylinder(ρ::Vector{Float64}, D::Vector{Float64}, μa::Vector{Float64}, a::Float64, zb::Vector{Float64}, z::ForwardDiff.Dual{ForwardDiff.Tag{LightPropagation.var"#69#71"{Vector{Float64}, Vector{Float64}, Vector{Float64}, Float64, Vector{Float64}, Vector{Float64}, Float64, Vector{Float64}}, Float64}, Float64, 1}, z0::Float64, l::Vector{Float64}, n_med::Vector{Float64}, besselroots::Vector{Float64}, green::typeof(LightPropagation._green_Nlaycylin_bottom), N::Int64)
   @ LightPropagation ~/.julia/packages/LightPropagation/fzSt3/src/forwardmodels/Diffusion Approximation/DAcylinder_layered.jl:322
 [4] fluence_DA_Nlay_cylinder_CW(ρ::Vector{Float64}, μa::Vector{Float64}, μsp::Vector{Float64}, n_ext::Float64, n_med::Vector{Float64}, l::Vector{Float64}, a::Float64, z::ForwardDiff.Dual{ForwardDiff.Tag{LightPropagation.var"#69#71"{Vector{Float64}, Vector{Float64}, Vector{Float64}, Float64, Vector{Float64}, Vector{Float64}, Float64, Vector{Float64}}, Float64}, Float64, 1}, besselroots::Vector{Float64})
   @ LightPropagation ~/.julia/packages/LightPropagation/fzSt3/src/forwardmodels/Diffusion Approximation/DAcylinder_layered.jl:63
 [5] #69
   @ ~/.julia/packages/LightPropagation/fzSt3/src/forwardmodels/Diffusion Approximation/DAcylinder_layered.jl:108 [inlined]
 [6] derivative
   @ ~/.julia/packages/ForwardDiff/5gUap/src/derivative.jl:14 [inlined]
 [7] flux_DA_Nlay_cylinder_CW(ρ::Vector{Float64}, μa::Vector{Float64}, μsp::Vector{Float64}, n_ext::Float64, n_med::Vector{Float64}, l::Vector{Float64}, a::Float64, z::Float64, besselroots::Vector{Float64})
   @ LightPropagation ~/.julia/packages/LightPropagation/fzSt3/src/forwardmodels/Diffusion Approximation/DAcylinder_layered.jl:108
 [8] flux_DA_Nlay_cylinder_CW(data::Nlayer_cylinder{Float64})
   @ LightPropagation ~/.julia/packages/LightPropagation/fzSt3/src/forwardmodels/Diffusion Approximation/DAcylinder_layered.jl:121
 [9] top-level scope
   @ REPL[8]:1

[heltonmc]

Want to specify this works when ρ is a scalar.

julia> data = Nlayer_cylinder(a = 10.0, l = [1.0, 1.0, 1.0, 2.0], z = 5.0, ρ = 1.0)
Nlayer_cylinder{Float64}

julia> flux_DA_Nlay_cylinder_CW(data)
4.342134601324081e-6

A work around is to just call the scalar version on many values of ρ

julia> map(ρ -> flux_DA_Nlay_cylinder_CW(Nlayer_cylinder(a = 10.0, ρ = ρ)), [0.5, 1.0, 1.5])
3-element Vector{Float64}:
 0.1298666669051858
 0.011815624951152489
 0.0020139880886783975

Though this does come at a computational hit because we are redoing a bunch of similar calculations. I am not actively working on fixing this right now. I am hoping to maybe transition to some of the new AD platforms in the future....

Edit: I will probably implement the analytical derivatives for this which will resolve this issue.