Move sparse AD code to SparseMatrixColorings extension

DifferentiationInterface/Project.toml

@@ -11,7 +11,6 @@ FillArrays = "1a297f60-69ca-5386-bcde-b61e274b549b"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
 PackageExtensionCompat = "65ce6f38-6b18-4e1d-a461-8949797d7930"
 SparseArrays = "2f01184e-e22b-5df5-ae63-d93ebab69eaf"
-SparseMatrixColorings = "0a514795-09f3-496d-8182-132a7b665d35"
 
 [weakdeps]
 ChainRulesCore = "d360d2e6-b24c-11e9-a2a3-2a2ae2dbcce4"
@@ -23,6 +22,7 @@ FiniteDifferences = "26cc04aa-876d-5657-8c51-4c34ba976000"
 ForwardDiff = "f6369f11-7733-5829-9624-2563aa707210"
 PolyesterForwardDiff = "98d1487c-24ca-40b6-b7ab-df2af84e126b"
 ReverseDiff = "37e2e3b7-166d-5795-8a7a-e32c996b4267"
+SparseMatrixColorings = "0a514795-09f3-496d-8182-132a7b665d35"
 Symbolics = "0c5d862f-8b57-4792-8d23-62f2024744c7"
 Tapir = "07d77754-e150-4737-8c94-cd238a1fb45b"
 Tracker = "9f7883ad-71c0-57eb-9f7f-b5c9e6d3789c"
@@ -38,6 +38,7 @@ DifferentiationInterfaceFiniteDifferencesExt = "FiniteDifferences"
 DifferentiationInterfaceForwardDiffExt = "ForwardDiff"
 DifferentiationInterfacePolyesterForwardDiffExt = "PolyesterForwardDiff"
 DifferentiationInterfaceReverseDiffExt = "ReverseDiff"
+DifferentiationInterfaceSparseMatrixColoringsExt = "SparseMatrixColorings"
 DifferentiationInterfaceSymbolicsExt = "Symbolics"
 DifferentiationInterfaceTapirExt = "Tapir"
 DifferentiationInterfaceTrackerExt = "Tracker"

DifferentiationInterface/ext/DifferentiationInterfaceSparseMatrixColoringsExt/DifferentiationInterfaceSparseMatrixColoringsExt.jl

@@ -0,0 +1,40 @@
+module DifferentiationInterfaceSparseMatrixColoringsExt
+
+using ADTypes: AutoSparse, dense_ad
+using ADTypes: coloring_algorithm, sparsity_detector, jacobian_sparsity, hessian_sparsity
+using Compat
+using DifferentiationInterface
+using DifferentiationInterface:
+    Batch,
+    GradientExtras,
+    JacobianExtras,
+    HessianExtras,
+    HVPExtras,
+    hvp_batched,
+    make_seed,
+    maybe_inner,
+    pick_batchsize,
+    pushforward_batched,
+    pullback_batched,
+    prepare_gradient,
+    prepare_hvp_batched_same_point,
+    prepare_pushforward_batched_same_point,
+    prepare_pullback_batched_same_point
+import DifferentiationInterface as DI
+using SparseMatrixColorings:
+    AbstractColoringResult,
+    ColoringProblem,
+    GreedyColoringAlgorithm,
+    coloring,
+    column_colors,
+    row_colors,
+    column_groups,
+    row_groups,
+    decompress,
+    decompress!
+
+include("fallbacks.jl")
+include("jacobian.jl")
+include("hessian.jl")
+
+end

DifferentiationInterface/ext/DifferentiationInterfaceSparseMatrixColoringsExt/fallbacks.jl

@@ -0,0 +1,9 @@
+DI.check_available(backend::AutoSparse) = DI.check_available(dense_ad(backend))
+DI.twoarg_support(backend::AutoSparse) = DI.twoarg_support(dense_ad(backend))
+
+function DI.pushforward_performance(backend::AutoSparse)
+    return DI.pushforward_performance(dense_ad(backend))
+end
+
+DI.pullback_performance(backend::AutoSparse) = DI.pullback_performance(dense_ad(backend))
+DI.hvp_mode(backend::AutoSparse{<:SecondOrder}) = DI.hvp_mode(dense_ad(backend))

DifferentiationInterface/src/sparse/hessian.jl → DifferentiationInterface/ext/DifferentiationInterfaceSparseMatrixColoringsExt/hessian.jl

@@ -35,7 +35,7 @@ end
 
 ## Hessian, one argument
 
-function prepare_hessian(f::F, backend::AutoSparse, x) where {F}
+function DI.prepare_hessian(f::F, backend::AutoSparse, x) where {F}
     dense_backend = dense_ad(backend)
     sparsity = hessian_sparsity(f, x, sparsity_detector(backend))
     problem = ColoringProblem{:symmetric,:column}()
@@ -65,7 +65,9 @@ function prepare_hessian(f::F, backend::AutoSparse, x) where {F}
     )
 end
 
-function hessian(f::F, backend::AutoSparse, x, extras::SparseHessianExtras{B}) where {F,B}
+function DI.hessian(
+    f::F, backend::AutoSparse, x, extras::SparseHessianExtras{B}
+) where {F,B}
     @compat (; coloring_result, batched_seeds, hvp_batched_extras) = extras
     dense_backend = dense_ad(backend)
     Ng = length(column_groups(coloring_result))
@@ -86,7 +88,7 @@ function hessian(f::F, backend::AutoSparse, x, extras::SparseHessianExtras{B}) w
     return decompress(compressed_matrix, coloring_result)
 end
 
-function hessian!(
+function DI.hessian!(
     f::F, hess, backend::AutoSparse, x, extras::SparseHessianExtras{B}
 ) where {F,B}
     @compat (;
@@ -125,7 +127,7 @@ function hessian!(
     return hess
 end
 
-function value_gradient_and_hessian!(
+function DI.value_gradient_and_hessian!(
     f::F, grad, hess, backend::AutoSparse, x, extras::SparseHessianExtras
 ) where {F}
     y, _ = value_and_gradient!(
@@ -135,7 +137,7 @@ function value_gradient_and_hessian!(
     return y, grad, hess
 end
 
-function value_gradient_and_hessian(
+function DI.value_gradient_and_hessian(
     f::F, backend::AutoSparse, x, extras::SparseHessianExtras
 ) where {F}
     y, grad = value_and_gradient(

DifferentiationInterface/src/sparse/jacobian.jl → DifferentiationInterface/ext/DifferentiationInterfaceSparseMatrixColoringsExt/jacobian.jl

@@ -64,16 +64,16 @@ function PullbackSparseJacobianExtras{B}(;
     )
 end
 
-function prepare_jacobian(f::F, backend::AutoSparse, x) where {F}
+function DI.prepare_jacobian(f::F, backend::AutoSparse, x) where {F}
     y = f(x)
     return prepare_sparse_jacobian_aux(
-        (f,), backend, x, y, pushforward_performance(backend)
+        (f,), backend, x, y, DI.pushforward_performance(backend)
     )
 end
 
-function prepare_jacobian(f!::F, y, backend::AutoSparse, x) where {F}
+function DI.prepare_jacobian(f!::F, y, backend::AutoSparse, x) where {F}
     return prepare_sparse_jacobian_aux(
-        (f!, y), backend, x, y, pushforward_performance(backend)
+        (f!, y), backend, x, y, DI.pushforward_performance(backend)
     )
 end
 
@@ -149,49 +149,51 @@ end
 
 ## One argument
 
-function jacobian(f::F, backend::AutoSparse, x, extras::SparseJacobianExtras) where {F}
+function DI.jacobian(f::F, backend::AutoSparse, x, extras::SparseJacobianExtras) where {F}
     return sparse_jacobian_aux((f,), backend, x, extras)
 end
 
-function jacobian!(
+function DI.jacobian!(
     f::F, jac, backend::AutoSparse, x, extras::SparseJacobianExtras
 ) where {F}
     return sparse_jacobian_aux!((f,), jac, backend, x, extras)
 end
 
-function value_and_jacobian(
+function DI.value_and_jacobian(
     f::F, backend::AutoSparse, x, extras::SparseJacobianExtras
 ) where {F}
     return f(x), jacobian(f, backend, x, extras)
 end
 
-function value_and_jacobian!(
+function DI.value_and_jacobian!(
     f::F, jac, backend::AutoSparse, x, extras::SparseJacobianExtras
 ) where {F}
     return f(x), jacobian!(f, jac, backend, x, extras)
 end
 
 ## Two arguments
 
-function jacobian(f!::F, y, backend::AutoSparse, x, extras::SparseJacobianExtras) where {F}
+function DI.jacobian(
+    f!::F, y, backend::AutoSparse, x, extras::SparseJacobianExtras
+) where {F}
     return sparse_jacobian_aux((f!, y), backend, x, extras)
 end
 
-function jacobian!(
+function DI.jacobian!(
     f!::F, y, jac, backend::AutoSparse, x, extras::SparseJacobianExtras
 ) where {F}
     return sparse_jacobian_aux!((f!, y), jac, backend, x, extras)
 end
 
-function value_and_jacobian(
+function DI.value_and_jacobian(
     f!::F, y, backend::AutoSparse, x, extras::SparseJacobianExtras
 ) where {F}
     jac = jacobian(f!, y, backend, x, extras)
     f!(y, x)
     return y, jac
 end
 
-function value_and_jacobian!(
+function DI.value_and_jacobian!(
     f!::F, y, jac, backend::AutoSparse, x, extras::SparseJacobianExtras
 ) where {F}
     jacobian!(f!, y, jac, backend, x, extras)

DifferentiationInterface/src/DifferentiationInterface.jl

@@ -12,8 +12,7 @@ module DifferentiationInterface
 using ADTypes: ADTypes, AbstractADType
 using ADTypes: mode, ForwardMode, ForwardOrReverseMode, ReverseMode, SymbolicMode
 using ADTypes: AutoSparse, dense_ad
-using ADTypes: coloring_algorithm
-using ADTypes: sparsity_detector, jacobian_sparsity, hessian_sparsity
+using ADTypes: coloring_algorithm, sparsity_detector, jacobian_sparsity, hessian_sparsity
 using ADTypes:
     AutoChainRules,
     AutoDiffractor,
@@ -33,18 +32,7 @@ using DocStringExtensions
 using FillArrays: OneElement
 using LinearAlgebra: Symmetric, Transpose, dot, parent, transpose
 using PackageExtensionCompat: @require_extensions
-using SparseArrays: SparseMatrixCSC, nonzeros, nzrange, rowvals, sparse
-using SparseMatrixColorings:
-    AbstractColoringResult,
-    ColoringProblem,
-    GreedyColoringAlgorithm,
-    coloring,
-    column_colors,
-    row_colors,
-    column_groups,
-    row_groups,
-    decompress,
-    decompress!
+using SparseArrays: sparse
 
 abstract type Extras end
 
@@ -73,10 +61,6 @@ include("second_order/hessian.jl")
 
 include("fallbacks/no_extras.jl")
 
-include("sparse/fallbacks.jl")
-include("sparse/jacobian.jl")
-include("sparse/hessian.jl")
-
 include("misc/differentiate_with.jl")
 include("misc/sparsity_detector.jl")
 include("misc/from_primitive.jl")