Relax orthogonal (#53)

* Relax to Abstract

* Bump Project

* Update docstring

Project.toml

@@ -1,7 +1,7 @@
 name = "ParameterHandling"
 uuid = "2412ca09-6db7-441c-8e3a-88d5709968c5"
 authors = ["Invenia Technical Computing Corporation"]
-version = "0.4.2"
+version = "0.4.3"
 
 [deps]
 ChainRulesCore = "d360d2e6-b24c-11e9-a2a3-2a2ae2dbcce4"

src/parameters_matrix.jl

@@ -1,18 +1,18 @@
 """
-    nearest_orthogonal_matrix(X::StridedMatrix)
+    nearest_orthogonal_matrix(X::AbstractMatrix{<:Union{Real,Complex}})
 
 Project `X` onto the closest orthogonal matrix in Frobenius norm.
 
 Originally used in varz: https://github.com/wesselb/varz/blob/master/varz/vars.py#L446
 """
-@inline function nearest_orthogonal_matrix(X::StridedMatrix{<:Union{Real,Complex}})
+@inline function nearest_orthogonal_matrix(X::AbstractMatrix{<:Union{Real,Complex}})
     # Inlining necessary for type inference for some reason.
     U, _, V = svd(X)
     return U * V'
 end
 
 """
-    orthogonal(X::StridedMatrix{<:Real})
+    orthogonal(X::AbstractMatrix{<:Real})
 
 Produce a parameter whose `value` is constrained to be an orthogonal matrix. The argument `X` need not
 be orthogonal.
@@ -22,9 +22,9 @@ Frobenius norm) and is overparametrised as a consequence.
 
 Originally used in varz: https://github.com/wesselb/varz/blob/master/varz/vars.py#L446
 """
-orthogonal(X::StridedMatrix{<:Real}) = Orthogonal(X)
+orthogonal(X::AbstractMatrix{<:Real}) = Orthogonal(X)
 
-struct Orthogonal{TX<:StridedMatrix{<:Real}} <: AbstractParameter
+struct Orthogonal{TX<:AbstractMatrix{<:Real}} <: AbstractParameter
     X::TX
 end
 
@@ -39,14 +39,14 @@ function flatten(::Type{T}, X::Orthogonal) where {T<:Real}
 end
 
 """
-    positive_definite(X::StridedMatrix{<:Real})
+    positive_definite(X::AbstractMatrix{<:Real})
 
 Produce a parameter whose `value` is constrained to be a positive-definite matrix. The argument `X` needs to
 be a positive-definite matrix (see https://en.wikipedia.org/wiki/Definite_matrix).
 
 The unconstrained parameter is a `LowerTriangular` matrix, stored as a vector.
 """
-function positive_definite(X::StridedMatrix{<:Real})
+function positive_definite(X::AbstractMatrix{<:Real})
     isposdef(X) || throw(ArgumentError("X is not positive-definite"))
     return PositiveDefinite(tril_to_vec(cholesky(X).L))
 end