relax type requirement on TwiceDifferentiable (#140)

* relax type requirement on TwiceDifferentiable

* added tests for ComponentArrays

Co-authored-by: amrods <[email protected]>

Author: longemen3000

Project.toml

@@ -15,6 +15,7 @@ FiniteDiff = "2.0"
 julia = "1.5"
 
 [extras]
+ComponentArrays = "b0b7db55-cfe3-40fc-9ded-d10e2dbeff66"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
 OptimTestProblems = "cec144fc-5a64-5bc6-99fb-dde8f63e154c"
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
@@ -23,4 +24,4 @@ SparseArrays = "2f01184e-e22b-5df5-ae63-d93ebab69eaf"
 Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
 
 [targets]
-test = ["LinearAlgebra", "OptimTestProblems", "Random", "RecursiveArrayTools", "SparseArrays", "Test"]
+test = ["ComponentArrays", "LinearAlgebra", "OptimTestProblems", "Random", "RecursiveArrayTools", "SparseArrays", "Test"]

src/objective_types/twicedifferentiable.jl

@@ -113,12 +113,12 @@ function TwiceDifferentiable(f, x::AbstractArray, F::Real = real(zero(eltype(x))
             FiniteDiff.finite_difference_gradient!(storage, f, x, gcache)
             return
         end
-        function fg!(storage::Vector, x::Vector)
+        function fg!(storage, x)
             g!(storage, x)
             return f(x)
         end
 
-        function h!(storage::Matrix, x::Vector)
+        function h!(storage, x)
             FiniteDiff.finite_difference_hessian!(storage, f, x)
             return
         end

test/abstractarrays.jl

@@ -1,3 +1,26 @@
+@testset "ComponentArrays" begin
+    x_seed_1 = [0.0]
+    x_seed_2 = [0.0]
+    x_seed = ComponentArray(x_seed_1=x_seed_1, x_seed_2=x_seed_2)
+    g_seed_1 = [0.0]
+    g_seed_2 = [0.0]
+    g_seed = ComponentArray(g_seed_1=g_seed_1, g_seed_2=g_seed_2)
+    f_x_seed = 8157.682077608529
+
+    nd = NonDifferentiable(exponential, x_seed)
+    @test nd.f == exponential
+    @test value(nd) == 0.0
+    @test nd.f_calls == [0]
+    od = OnceDifferentiable(exponential, exponential_gradient!, nothing, x_seed, 0.0, g_seed)
+    @test od.f == exponential
+    @test od.df == exponential_gradient!
+    @test value(od) == 0.0
+    @test od.f_calls == [0]
+    @test od.df_calls == [0]
+    @test typeof(od.DF) <: ComponentArray
+    @test typeof(od.x_f) <: ComponentArray
+    @test typeof(od.x_df) <: ComponentArray
+end
 @testset "Matrix OnceDifferentiable" begin
     x_seed = fill(0.0, 1, 2)
     g_seed = fill(0.0, 1, 2)

test/runtests.jl

@@ -1,6 +1,7 @@
 using NLSolversBase, Test
 using Random
 using LinearAlgebra: Diagonal, I
+using ComponentArrays
 using SparseArrays
 using OptimTestProblems
 using RecursiveArrayTools