Reexport backends from ADTypes (#217)

* Reexport ADTypes

* Remove ADTypes namespacing when possible

* Include reexported ADTypes in API reference

* Minor fixes

* Move backend API reference to "Backends" page

Author: adrhill

DifferentiationInterface/docs/make.jl

@@ -4,7 +4,7 @@ import DifferentiationInterface as DI
 using Documenter
 using DocumenterMermaid
 
-using ADTypes
+using ADTypes: ADTypes
 using Diffractor: Diffractor
 using Enzyme: Enzyme
 using FastDifferentiation: FastDifferentiation
@@ -37,16 +37,18 @@ extensions = [
 ]
 
 makedocs(;
-    modules=[DifferentiationInterface, extensions...],
+    modules=[DifferentiationInterface, ADTypes, extensions...],
     authors="Guillaume Dalle, Adrian Hill",
     sitename="DifferentiationInterface.jl",
     format=Documenter.HTML(),
     pages=[
-        "Home" => "index.md", #
+        "Home" => "index.md",
         "Start here" => ["tutorial.md", "overview.md", "backends.md"],
         "API reference" => "api.md",
         "Advanced" => ["design.md", "extensions.md", "overloads.md"],
     ],
+    checkdocs=:exports,
+    warnonly=[:missing_docs, :cross_references],
 )
 
 deploydocs(;

DifferentiationInterface/docs/src/backends.md

@@ -4,7 +4,6 @@ CollapsedDocStrings = true
 ```
 
 ```@setup backends
-using ADTypes
 using DifferentiationInterface
 using DifferentiationInterface: backend_string
 import Markdown
@@ -70,3 +69,27 @@ You can check this compatibility using [`check_twoarg`](@ref).
 
 Only some backends are able to compute Hessians.
 You can use [`check_hessian`](@ref) to check this feature.
+
+
+## API Reference
+
+!!! warning
+    The following documentation has been re-exported from [ADTypes.jl](https://github.com/SciML/ADTypes.jl).
+    Refer to the ADTypes documentation for more information.
+
+```@docs
+AutoChainRules
+AutoDiffractor
+AutoEnzyme
+AutoFastDifferentiation
+AutoFiniteDiff
+AutoFiniteDifferences
+AutoForwardDiff
+AutoPolyesterForwardDiff
+AutoReverseDiff
+AutoSymbolics
+AutoTapir
+AutoTracker
+AutoZygote
+AutoSparse
+```

DifferentiationInterface/docs/src/overloads.md

@@ -22,7 +22,7 @@ Each cell can have three values:
     Check marks (✅) are clickable and link to the source code.
 
 ```@setup overloads
-using ADTypes
+using ADTypes: AbstractADType
 using DifferentiationInterface
 using DifferentiationInterface: backend_string, mutation_support, MutationSupported
 using Markdown: Markdown

DifferentiationInterface/src/DifferentiationInterface.jl

@@ -87,4 +87,21 @@ export prepare_second_derivative, prepare_hvp, prepare_hessian
 
 export check_available, check_twoarg, check_hessian
 
+# Re-export backends from ADTypes
+export AutoChainRules
+export AutoDiffractor
+export AutoEnzyme
+export AutoFastDifferentiation
+export AutoFiniteDiff
+export AutoFiniteDifferences
+export AutoForwardDiff
+export AutoPolyesterForwardDiff
+export AutoReverseDiff
+export AutoSymbolics
+export AutoTapir
+export AutoTracker
+export AutoZygote
+
+export AutoSparse
+
 end # module

DifferentiationInterface/src/utils/printing.jl

@@ -1,4 +1,5 @@
 backend_package_name(b::AbstractADType) = strip(string(b), ['(', ')'])
+backend_package_name(b::AutoSparse) = backend_package_name(dense_ad(b))
 
 backend_package_name(::AutoChainRules) = "ChainRules"
 backend_package_name(::AutoDiffractor) = "Diffractor"

DifferentiationInterfaceTest/docs/src/tutorial.md

@@ -8,7 +8,7 @@ We present a typical workflow with DifferentiationInterfaceTest.jl, building on
 
 ```@repl tuto
 using DifferentiationInterface, DifferentiationInterfaceTest
-import ADTypes, ForwardDiff, Enzyme
+import ForwardDiff, Enzyme
 import DataFrames, Markdown, PrettyTables, Printf
 ```
 
@@ -17,7 +17,7 @@ import DataFrames, Markdown, PrettyTables, Printf
 The AD backends we want to compare are [ForwardDiff.jl](https://github.com/JuliaDiff/ForwardDiff.jl) and [Enzyme.jl](https://github.com/EnzymeAD/Enzyme.jl).
 
 ```@repl tuto
-backends = [ADTypes.AutoForwardDiff(), ADTypes.AutoEnzyme(; mode=Enzyme.Reverse)]
+backends = [AutoForwardDiff(), AutoEnzyme(; mode=Enzyme.Reverse)]
 ```
 
 To do that, we are going to take gradients of a simple function: