Merge pull request #72 from SciML/gd/constant_f_enzyme

Add constant_function kwarg to AutoEnzyme

Project.toml

@@ -3,7 +3,7 @@ uuid = "47edcb42-4c32-4615-8424-f2b9edc5f35b"
 authors = [
     "Vaibhav Dixit <[email protected]>, Guillaume Dalle and contributors",
 ]
-version = "1.5.4"
+version = "1.6.0"
 
 [deps]
 ChainRulesCore = "d360d2e6-b24c-11e9-a2a3-2a2ae2dbcce4"

src/dense.jl

@@ -39,25 +39,63 @@ struct AutoDiffractor <: AbstractADType end
 mode(::AutoDiffractor) = ForwardOrReverseMode()
 
 """
-    AutoEnzyme{M}
+    AutoEnzyme{M,constant_function}
 
 Struct used to select the [Enzyme.jl](https://github.com/EnzymeAD/Enzyme.jl) backend for automatic differentiation.
 
 Defined by [ADTypes.jl](https://github.com/SciML/ADTypes.jl).
 
 # Constructors
 
-    AutoEnzyme(; mode=nothing)
+    AutoEnzyme(; mode=nothing, constant_function::Bool=false)
+
+The `constant_function` keyword argument (and type parameter) determines whether the function object itself should be considered constant or not during differentiation with Enzyme.jl.
+For simple functions, `constant_function` should usually be set to `false`, but in the case of closures or callable structs which contain differentiated data that can be treated as constant, `constant_function` should be set to `true` for increased performance (more details below).
 
 # Fields
 
   - `mode::M`: can be either
 
       + an object subtyping `EnzymeCore.Mode` (like `EnzymeCore.Forward` or `EnzymeCore.Reverse`) if a specific mode is required
       + `nothing` to choose the best mode automatically
+
+# Notes
+
+If `constant_function = true` but the enclosed data is not truly constant, then Enzyme.jl will not compute the correct derivative values.
+An example of such a function is:
+
+```julia
+cache = [0.0]
+function f(x)
+    cache[1] = x[1]^2
+    cache[1] + x[1]
+end
+```
+
+In this case, the enclosed cache is a function of the differentiated input, and thus its values are non-constant with respect to the input.
+Thus, in order to compute the correct derivative of the output, the derivative must propagate through the `cache` value, and said `cache` must not be treated as constant.
+
+Conversely, the following function can treat `parameter` as a constant, because `parameter` is never modified based on the input `x`:
+
+```julia
+parameter = [0.0]
+function f(x)
+    parameter[1] + x[1]
+end
+```
+
+In this case, `constant_function = true` would allow the chosen differentiation system to perform extra memory and compute optimizations, under the assumption that `parameter` is kept constant.
 """
-Base.@kwdef struct AutoEnzyme{M} <: AbstractADType
-    mode::M = nothing
+struct AutoEnzyme{M, constant_function} <: AbstractADType
+    mode::M
+end
+
+function AutoEnzyme(mode::M; constant_function::Bool = false) where {M}
+    return AutoEnzyme{M, constant_function}(mode)
+end
+
+function AutoEnzyme(; mode::M = nothing, constant_function::Bool = false) where {M}
+    return AutoEnzyme{M, constant_function}(mode)
 end
 
 mode(::AutoEnzyme) = ForwardOrReverseMode()  # specialized in the extension

test/dense.jl

@@ -28,19 +28,25 @@ end
 @testset "AutoEnzyme" begin
     ad = AutoEnzyme()
     @test ad isa AbstractADType
-    @test ad isa AutoEnzyme{Nothing}
+    @test ad isa AutoEnzyme{Nothing, false}
     @test mode(ad) isa ForwardOrReverseMode
     @test ad.mode === nothing
 
+    ad = AutoEnzyme(EnzymeCore.Forward; constant_function = true)
+    @test ad isa AbstractADType
+    @test ad isa AutoEnzyme{typeof(EnzymeCore.Forward), true}
+    @test mode(ad) isa ForwardMode
+    @test ad.mode == EnzymeCore.Forward
+
     ad = AutoEnzyme(; mode = EnzymeCore.Forward)
     @test ad isa AbstractADType
-    @test ad isa AutoEnzyme{typeof(EnzymeCore.Forward)}
+    @test ad isa AutoEnzyme{typeof(EnzymeCore.Forward), false}
     @test mode(ad) isa ForwardMode
     @test ad.mode == EnzymeCore.Forward
 
-    ad = AutoEnzyme(; mode = EnzymeCore.Reverse)
+    ad = AutoEnzyme(; mode = EnzymeCore.Reverse, constant_function = true)
     @test ad isa AbstractADType
-    @test ad isa AutoEnzyme{typeof(EnzymeCore.Reverse)}
+    @test ad isa AutoEnzyme{typeof(EnzymeCore.Reverse), true}
     @test mode(ad) isa ReverseMode
     @test ad.mode == EnzymeCore.Reverse
 end

test/misc.jl

@@ -21,11 +21,6 @@ end
     @test length(string(sparse_backend1)) < length(string(sparse_backend2))
 end
 
-import ADTypes
-
-struct FakeSparsityDetector <: ADTypes.AbstractSparsityDetector end
-struct FakeColoringAlgorithm <: ADTypes.AbstractColoringAlgorithm end
-
 for backend in [
     # dense
     ADTypes.AutoChainRules(; ruleconfig = :rc),