Add Enzyme rules for factorizations

.github/workflows/CI.yml

@@ -22,6 +22,6 @@ jobs:
     with:
       fast: "${{ github.event.pull_request.draft == true }}"
       exclude: '["cuda", "amd"]'
-      timeout-minutes: 120
+      timeout-minutes: 150
     secrets:
       CODECOV_TOKEN: ${{ secrets.CODECOV_TOKEN }}

Project.toml

@@ -65,10 +65,13 @@ Preferences = "1"
 Printf = "1"
 Random = "1"
 ScopedValues = "1.3.0"
-Strided = "2.6.1"
+Strided = "2.6.3"
 TensorKitSectors = "0.3.7"
 TensorOperations = "5.5.2, 5.6"
 TimerOutputs = "1"
 TupleTools = "1.5"
 VectorInterface = "0.6"
 julia = "1.10"
+
+[sources]
+MatrixAlgebraKit = {url = "https://github.com/QuantumKitHub/MatrixAlgebraKit.jl/", rev = "main"}

ext/TensorKitEnzymeExt/TensorKitEnzymeExt.jl

@@ -15,5 +15,6 @@ include("utility.jl")
 include("linalg.jl")
 include("indexmanipulations.jl")
 include("tensoroperations.jl")
+include("factorizations.jl")
 
 end

ext/TensorKitEnzymeExt/factorizations.jl

@@ -0,0 +1,69 @@
+# need these due to Enzyme choking on blocks
+
+for f in (:project_hermitian, :project_antihermitian)
+    f! = Symbol(f, :!)
+    @eval begin
+        function EnzymeRules.augmented_primal(
+                config::EnzymeRules.RevConfigWidth{1},
+                func::Const{typeof($f!)},
+                ::Type{RT},
+                A::Annotation{<:AbstractTensorMap},
+                arg::Annotation{<:AbstractTensorMap},
+                alg::Const,
+            ) where {RT}
+            $f!(A.val, arg.val, alg.val)
+            primal = EnzymeRules.needs_primal(config) ? arg.val : nothing
+            shadow = EnzymeRules.needs_shadow(config) ? arg.dval : nothing
+            cache = nothing
+            return EnzymeRules.AugmentedReturn(primal, shadow, cache)
+        end
+        function EnzymeRules.reverse(
+                config::EnzymeRules.RevConfigWidth{1},
+                func::Const{typeof($f!)},
+                ::Type{RT},
+                cache,
+                A::Annotation{<:AbstractTensorMap},
+                arg::Annotation{<:AbstractTensorMap},
+                alg::Const,
+            ) where {RT}
+            if !isa(A, Const) && !isa(arg, Const)
+                $f!(arg.dval, arg.dval, alg.val)
+                if A.dval !== arg.dval
+                    A.dval .+= arg.dval
+                    make_zero!(arg.dval)
+                end
+            end
+            return (nothing, nothing, nothing)
+        end
+        function EnzymeRules.augmented_primal(
+                config::EnzymeRules.RevConfigWidth{1},
+                func::Const{typeof($f)},
+                ::Type{RT},
+                A::Annotation{<:AbstractTensorMap},
+                alg::Const,
+            ) where {RT}
+            ret = $f(A.val, alg.val)
+            dret = make_zero(ret)
+            primal = EnzymeRules.needs_primal(config) ? ret : nothing
+            shadow = EnzymeRules.needs_shadow(config) ? dret : nothing
+            cache = dret
+            return EnzymeRules.AugmentedReturn(primal, shadow, cache)
+        end
+        function EnzymeRules.reverse(
+                config::EnzymeRules.RevConfigWidth{1},
+                func::Const{typeof($f)},
+                ::Type{RT},
+                cache,
+                A::Annotation{<:AbstractTensorMap},
+                alg::Const,
+            ) where {RT}
+            dret = cache
+            if !isa(A, Const)
+                $f!(dret, dret, alg.val)
+                add!(A.dval, dret)
+            end
+            make_zero!(dret)
+            return (nothing, nothing)
+        end
+    end
+end

ext/TensorKitEnzymeExt/utility.jl

@@ -148,6 +148,7 @@ end
 @inline EnzymeRules.inactive(::typeof(TensorKit.insertleftunit), ::HomSpace, ::Any) = nothing
 @inline EnzymeRules.inactive(::typeof(TensorKit.insertrightunit), ::HomSpace, ::Any) = nothing
 @inline EnzymeRules.inactive(::typeof(TensorKit.removeunit), ::HomSpace, ::Any) = nothing
+@inline EnzymeRules.inactive(::typeof(TensorKit.infimum), ::Any, ::Any) = nothing
 @inline EnzymeRules.inactive(::typeof(TensorKit.sectorstructure), ::Any) = nothing
 @inline EnzymeRules.inactive(::typeof(TensorKit.degeneracystructure), ::Any) = nothing
 @inline EnzymeRules.inactive(::typeof(TensorKit.select), s::HomSpace, i::Index2Tuple) = nothing

ext/TensorKitEnzymeTestUtilsExt.jl

@@ -43,6 +43,33 @@ function EnzymeTestUtils.to_vec(t::TensorKit.DiagonalTensorMap, seen_vecs::Enzym
     parent_vec, parent_t = to_vec(TensorMap(t), seen_vecs)
     return parent_vec, TensorKit.DiagonalTensorMap ∘ parent_t
 end
+function EnzymeTestUtils.to_vec(v::TensorKit.SectorVector, seen_vecs::EnzymeTestUtils.AliasDict)
+    has_seen = haskey(seen_vecs, v)
+    is_const = Enzyme.Compiler.guaranteed_const(Core.Typeof(v))
+    if has_seen || is_const
+        v_vec = Float32[]
+    else
+        vec_of_vecs = [b * TensorKit.sqrtdim(c) for (c, b) in pairs(v)]
+        v_vec, back = to_vec(vec_of_vecs)
+        seen_vecs[v] = v_vec
+    end
+    function SectorVector_from_vec(v_vec_new::AbstractVector, seen_xs::EnzymeTestUtils.AliasDict)
+        if xor(has_seen, haskey(seen_xs, v))
+            throw(ErrorException("Arrays must be reconstructed in the same order as they are vectorized."))
+        end
+        has_seen && return seen_xs[v]
+        is_const && return v
+
+        v_new = similar(v)
+        vvec_of_vecs = back(v_vec_new)
+        for (i, (c, b)) in enumerate(pairs(v_new))
+            scale!(b, vvec_of_vecs[i], TensorKit.invsqrtdim(c))
+        end
+        seen_xs[v] = v_new
+        return v_new
+    end
+    return v_vec, SectorVector_from_vec
+end
 
 # generate random tangents for testing
 function EnzymeTestUtils.rand_tangent(rng, t::TensorMap)

src/factorizations/diagonal.jl

@@ -3,6 +3,7 @@
 _repack_diagonal(d::DiagonalTensorMap) = Diagonal(d.data)
 _repack_diagonal(d::SectorVector) = Diagonal(parent(d))
 
+MAK.diagonal(t::SectorVector) = DiagonalTensorMap(t)
 MAK.diagview(t::DiagonalTensorMap) = SectorVector(t.data, TensorKit.diagonalblockstructure(space(t)))
 
 for f in (

src/factorizations/factorizations.jl

@@ -32,6 +32,7 @@ include("truncation.jl")
 include("adjoint.jl")
 include("diagonal.jl")
 include("pullbacks.jl")
+include("pushforwards.jl")
 
 TensorKit.one!(A::AbstractMatrix) = MatrixAlgebraKit.one!(A)
 

src/factorizations/pullbacks.jl

@@ -11,6 +11,16 @@ for pullback! in (
         end
         return Δt
     end
+    @eval function MAK.$pullback!(
+            Δt::AbstractTensorMap, ::Nothing, F, ΔF; kwargs...
+        )
+        foreachblock(Δt) do c, (Δb,)
+            Fc = block.(F, Ref(c))
+            ΔFc = block.(ΔF, Ref(c))
+            return MAK.$pullback!(Δb, nothing, Fc, ΔFc; kwargs...)
+        end
+        return Δt
+    end
 end
 for pullback! in (:qr_null_pullback!, :lq_null_pullback!)
     @eval function MAK.$pullback!(
@@ -26,21 +36,45 @@ for pullback! in (:qr_null_pullback!, :lq_null_pullback!)
 end
 _notrunc_ind(t) = SectorDict(c => Colon() for c in blocksectors(t))
 
+for pullback! in (:eig_vals_pullback!, :eigh_vals_pullback!)
+    @eval function MAK.$pullback!(
+            Δt::AbstractTensorMap, ::Nothing, DV::Tuple{Diagonal, <:AbstractTensorMap}, ΔD, inds;
+            kwargs...
+        )
+        return MAK.$pullback!(Δt, nothing, (MAK.diagonal(parent(DV[1])), DV[2]), ΔD, inds; kwargs...)
+    end
+end
+function MAK.svd_vals_pullback!(
+        Δt::AbstractTensorMap, ::Nothing, USVᴴ::Tuple{<:AbstractTensorMap, Diagonal, <:AbstractTensorMap}, ΔS, ind;
+        kwargs...
+    )
+    return MAK.svd_vals_pullback!(Δt, nothing, (USVᴴ[1], MAK.diagonal(parent(USVᴴ[2])), USVᴴ[3]), ΔS, ind; kwargs...)
+end
+
+nothing_or_block(x, c) = isnothing(x) ? x : block(x, c)
+nothing_or_block(x::Diagonal, c) = block(MAK.diagonal(parent(x)), c)
+nothing_or_foreachblock(f, Δt, t) = isnothing(t) ? foreachblock(f, Δt) : foreachblock(f, Δt, t)
 for pullback! in (:svd_pullback!, :eig_pullback!, :eigh_pullback!)
     @eval function MAK.$pullback!(
-            Δt::AbstractTensorMap, t::AbstractTensorMap, F, ΔF, inds = _notrunc_ind(t);
+            Δt::AbstractTensorMap, t, F, ΔF, inds = _notrunc_ind(Δt);
             kwargs...
         )
-        foreachblock(Δt, t) do c, (Δb, b)
+        nothing_or_foreachblock(Δt, t) do c, Δbb
+            Δb, b = length(Δbb) == 1 ? (only(Δbb), nothing) : Δbb
             ind = get(inds, c, nothing)
             isnothing(ind) && return nothing
-            Fc = block.(F, Ref(c))
-            ΔFc = block.(ΔF, Ref(c))
+            Fc = nothing_or_block.(F, Ref(c))
+            ΔFc = nothing_or_block.(ΔF, Ref(c))
             MAK.$pullback!(Δb, b, Fc, ΔFc, ind; kwargs...)
             return nothing
         end
         return Δt
     end
+    @eval function MAK.$pullback!(
+            Δt::AbstractTensorMap, t, F, ΔF, ::Colon; kwargs...
+        )
+        return MAK.$pullback!(Δt, t, F, ΔF, _notrunc_ind(Δt); kwargs...)
+    end
 end
 
 for pullback_trunc! in (:svd_trunc_pullback!, :eig_trunc_pullback!, :eigh_trunc_pullback!)
@@ -97,3 +131,7 @@ function MAK.remove_svd_gauge_dependence!(
     end
     return ΔU, ΔVᴴ
 end
+
+MAK.has_equal_storage(A::AbstractTensorMap, B::AbstractTensorMap) = A === B
+MAK.has_equal_storage(A::AbstractTensorMap, B::SectorVector) = false
+MAK.has_equal_storage(A::SectorVector, B::AbstractTensorMap) = false

src/factorizations/pushforwards.jl

@@ -0,0 +1,66 @@
+for pushforward! in (
+        :qr_pushforward!, :lq_pushforward!, :left_polar_pushforward!, :right_polar_pushforward!,
+    )
+    @eval function MAK.$pushforward!(
+            Δt::AbstractTensorMap, t::AbstractTensorMap, F, ΔF; kwargs...
+        )
+        foreachblock(Δt, t) do c, (Δb, b)
+            Fc = block.(F, Ref(c))
+            ΔFc = block.(ΔF, Ref(c))
+            return MAK.$pushforward!(Δb, b, Fc, ΔFc; kwargs...)
+        end
+        return Δt
+    end
+    @eval function MAK.$pushforward!(
+            Δt::AbstractTensorMap, ::Nothing, F, ΔF; kwargs...
+        )
+        foreachblock(Δt) do c, (Δb,)
+            Fc = block.(F, Ref(c))
+            ΔFc = block.(ΔF, Ref(c))
+            return MAK.$pushforward!(Δb, nothing, Fc, ΔFc; kwargs...)
+        end
+        return Δt
+    end
+end
+for pushforward! in (:qr_null_pushforward!, :lq_null_pushforward!)
+    @eval function MAK.$pushforward!(
+            Δt::AbstractTensorMap, t::AbstractTensorMap, F, ΔF; kwargs...
+        )
+        foreachblock(Δt, t) do c, (Δb, b)
+            Fc = block(F, c)
+            ΔFc = block(ΔF, c)
+            return MAK.$pushforward!(Δb, b, Fc, ΔFc; kwargs...)
+        end
+        return Δt
+    end
+end
+
+for pushforward! in (:eig_vals_pushforward!, :eigh_vals_pushforward!)
+    @eval function MAK.$pushforward!(
+            Δt::AbstractTensorMap, ::Nothing, DV::Tuple{Diagonal, <:AbstractTensorMap}, ΔD, inds;
+            kwargs...
+        )
+        return MAK.$pushforward!(Δt, nothing, (MAK.diagonal(parent(DV[1])), DV[2]), ΔD, inds; kwargs...)
+    end
+end
+function MAK.svd_vals_pushforward!(
+        Δt::AbstractTensorMap, ::Nothing, USVᴴ::Tuple{<:AbstractTensorMap, Diagonal, <:AbstractTensorMap}, ΔS, ind;
+        kwargs...
+    )
+    return MAK.svd_vals_pushforward!(Δt, nothing, (USVᴴ[1], MAK.diagonal(parent(USVᴴ[2])), USVᴴ[3]), ΔS, ind; kwargs...)
+end
+
+for pushforward! in (:svd_pushforward!, :eig_pushforward!, :eigh_pushforward!)
+    @eval function MAK.$pushforward!(
+            Δt::AbstractTensorMap, t, F, ΔF;
+            kwargs...
+        )
+        foreachblock(Δt, t) do c, (Δb, b)
+            Fc = nothing_or_block.(F, Ref(c))
+            ΔFc = nothing_or_block.(ΔF, Ref(c))
+            MAK.$pushforward!(Δb, b, Fc, ΔFc; kwargs...)
+            return nothing
+        end
+        return Δt
+    end
+end

test/enzyme-factorizations/eig.jl

@@ -0,0 +1,37 @@
+using Test, TestExtras
+using TensorKit
+using TensorOperations
+using MatrixAlgebraKit
+using MatrixAlgebraKit: remove_eig_gauge_dependence!
+using Enzyme, EnzymeTestUtils
+using Random
+
+is_ci = get(ENV, "CI", "false") == "true"
+
+spacelist = ad_spacelist(fast_tests)
+eltypes = (Float64, ComplexF64)
+
+@timedtestset "Enzyme - Factorizations (EIG): $(TensorKit.type_repr(sectortype(eltype(V)))) ($T)" for V in spacelist, T in eltypes, t in (randn(T, V[1] ← V[1]), rand(T, V[1] ⊗ V[2] ← V[1] ⊗ V[2]))
+    atol = default_tol(T)
+    rtol = default_tol(T)
+
+    if !is_ci
+        DV = eig_full(t)
+        ΔDV = EnzymeTestUtils.rand_tangent(DV)
+        remove_eig_gauge_dependence!(ΔDV[2], DV...)
+        EnzymeTestUtils.test_reverse(eig_full, Duplicated, (t, Duplicated); output_tangent = ΔDV, atol, rtol)
+        EnzymeTestUtils.test_forward(eig_full, Duplicated, (t, Duplicated); atol, rtol)
+
+        D = eig_vals(t)
+        EnzymeTestUtils.test_reverse(eig_vals, Duplicated, (t, Duplicated); atol, rtol)
+        EnzymeTestUtils.test_forward(eig_vals, Duplicated, (t, Duplicated); atol, rtol)
+    end
+
+    V_trunc = spacetype(t)(c => min(size(b)...) ÷ 2 for (c, b) in blocks(t))
+    trunc = truncspace(V_trunc)
+    alg = MatrixAlgebraKit.select_algorithm(eig_trunc_no_error, t, nothing; trunc)
+    DVtrunc = eig_trunc_no_error(t, alg)
+    ΔDVtrunc = EnzymeTestUtils.rand_tangent(DVtrunc)
+    remove_eig_gauge_dependence!(ΔDVtrunc[2], DVtrunc...)
+    EnzymeTestUtils.test_reverse(eig_trunc_no_error, Duplicated, (t, Duplicated), (alg, Const); output_tangent = ΔDVtrunc, atol, rtol)
+end

test/enzyme-factorizations/eigh.jl

@@ -0,0 +1,39 @@
+using Test, TestExtras
+using TensorKit
+using TensorOperations
+using MatrixAlgebraKit
+using MatrixAlgebraKit: remove_eigh_gauge_dependence!
+using Enzyme, EnzymeTestUtils
+using Random
+
+is_ci = get(ENV, "CI", "false") == "true"
+
+spacelist = ad_spacelist(fast_tests)
+eltypes = (Float64, ComplexF64)
+
+@timedtestset "Enzyme - Factorizations (EIGH): $(TensorKit.type_repr(sectortype(eltype(V)))) ($T)" for V in spacelist, T in eltypes, t in (randn(T, V[1] ← V[1]), rand(T, V[1] ⊗ V[2] ← V[1] ⊗ V[2]))
+    atol = default_tol(T)
+    rtol = default_tol(T)
+    th = project_hermitian(t)
+    if !is_ci
+        DV = eigh_full(th)
+        ΔDV = EnzymeTestUtils.rand_tangent(DV)
+        remove_eigh_gauge_dependence!(ΔDV[2], DV...)
+        proj_eigh_full(t) = eigh_full(project_hermitian(t))
+        EnzymeTestUtils.test_reverse(proj_eigh_full, Duplicated, (th, Duplicated); output_tangent = ΔDV, atol, rtol)
+        EnzymeTestUtils.test_forward(proj_eigh_full, Duplicated, (th, Duplicated); atol, rtol)
+
+        D = eigh_vals(th)
+        EnzymeTestUtils.test_reverse(eigh_vals ∘ project_hermitian, Duplicated, (th, Duplicated); atol, rtol)
+        EnzymeTestUtils.test_forward(eigh_vals ∘ project_hermitian, Duplicated, (th, Duplicated); atol, rtol)
+    end
+
+    V_trunc = spacetype(th)(c => min(size(b)...) ÷ 2 for (c, b) in blocks(t))
+    trunc = truncspace(V_trunc)
+    alg = MatrixAlgebraKit.select_algorithm(eigh_trunc_no_error, th, nothing; trunc)
+    DVtrunc = eigh_trunc_no_error(th, alg)
+    ΔDVtrunc = EnzymeTestUtils.rand_tangent(DVtrunc)
+    remove_eigh_gauge_dependence!(ΔDVtrunc[2], DVtrunc...)
+    proj_eigh(t, alg) = eigh_trunc_no_error(project_hermitian(t), alg)
+    EnzymeTestUtils.test_reverse(proj_eigh, Duplicated, (th, Duplicated), (alg, Const); output_tangent = ΔDVtrunc, atol, rtol)
+end