Fix test error with Turing >= 0.30 (#570)

* Fix test error with Turing >= 0.30

* Fix format

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* Fix doctests and format

* Fix LKJ tests

* Remove `setchunksize`

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

Author: devmotion

benchmarks/README.md

@@ -1,4 +1,5 @@
 To run the benchmarks, simply do:
+
 ```sh
 julia --project -e 'using DynamicPPLBenchmarks; weave_benchmarks();'
 ```

benchmarks/benchmark_body.jmd

@@ -20,7 +20,9 @@ results["evaluation_typed"]
 ```
 
 ```julia; echo=false; results="hidden";
-BenchmarkTools.save(joinpath("results", WEAVE_ARGS[:name], "$(nameof(m))_benchmarks.json"), results)
+BenchmarkTools.save(
+    joinpath("results", WEAVE_ARGS[:name], "$(nameof(m))_benchmarks.json"), results
+)
 ```
 
 ```julia; wrap=false
@@ -32,15 +34,16 @@ end
 ```julia; echo=false; results="hidden"
 if WEAVE_ARGS[:include_typed_code]
     # Serialize the output of `typed_code` so we can compare later.
-    haskey(WEAVE_ARGS, :name) && serialize(joinpath("results", WEAVE_ARGS[:name],"$(nameof(m)).jls"), string(typed));
+    haskey(WEAVE_ARGS, :name) &&
+        serialize(joinpath("results", WEAVE_ARGS[:name], "$(nameof(m)).jls"), string(typed))
 end
 ```
 
 ```julia; wrap=false; echo=false;
 if haskey(WEAVE_ARGS, :name_old)
     # We want to compare the generated code to the previous version.
-    import DiffUtils
-    typed_old = deserialize(joinpath("results", WEAVE_ARGS[:name_old], "$(nameof(m)).jls"));
-    DiffUtils.diff(typed_old, string(typed), width=130)
+    using DiffUtils: DiffUtils
+    typed_old = deserialize(joinpath("results", WEAVE_ARGS[:name_old], "$(nameof(m)).jls"))
+    DiffUtils.diff(typed_old, string(typed); width=130)
 end
 ```

benchmarks/benchmarks.jmd

@@ -19,21 +19,21 @@ import DynamicPPLBenchmarks: time_model_def, make_suite, typed_code, weave_child
     m ~ Normal()
     x ~ Normal(m, 1)
 
-    return (m = m, x = x)
+    return (m=m, x=x)
 end
 
 model_def = demo1;
 data = 1.0;
 ```
 
 ```julia; results="markup"; echo=false
-weave_child(WEAVE_ARGS[:benchmarkbody], mod = @__MODULE__, args = WEAVE_ARGS)
+weave_child(WEAVE_ARGS[:benchmarkbody]; mod=@__MODULE__, args=WEAVE_ARGS)
 ```
 
 ### `demo2`
 
 ```julia
-@model function demo2(y) 
+@model function demo2(y)
     # Our prior belief about the probability of heads in a coin.
     p ~ Beta(1, 1)
 
@@ -50,7 +50,7 @@ data = rand(0:1, 10);
 ```
 
 ```julia; results="markup"; echo=false
-weave_child(WEAVE_ARGS[:benchmarkbody], mod = @__MODULE__, args = WEAVE_ARGS)
+weave_child(WEAVE_ARGS[:benchmarkbody]; mod=@__MODULE__, args=WEAVE_ARGS)
 ```
 
 ### `demo3`
@@ -74,7 +74,7 @@ weave_child(WEAVE_ARGS[:benchmarkbody], mod = @__MODULE__, args = WEAVE_ARGS)
     k = Vector{Int}(undef, N)
     for i in 1:N
         k[i] ~ Categorical(w)
-        x[:,i] ~ MvNormal([μ[k[i]], μ[k[i]]], 1.)
+        x[:, i] ~ MvNormal([μ[k[i]], μ[k[i]]], 1.0)
     end
     return k
 end
@@ -88,11 +88,11 @@ N = 30
 μs = [-3.5, 0.0]
 
 # Construct the data points.
-data = mapreduce(c -> rand(MvNormal([μs[c], μs[c]], 1.), N), hcat, 1:2);
+data = mapreduce(c -> rand(MvNormal([μs[c], μs[c]], 1.0), N), hcat, 1:2);
 ```
 
 ```julia; echo=false
-weave_child(WEAVE_ARGS[:benchmarkbody], mod = @__MODULE__, args = WEAVE_ARGS)
+weave_child(WEAVE_ARGS[:benchmarkbody]; mod=@__MODULE__, args=WEAVE_ARGS)
 ```
 
 ### `demo4`: loads of indexing
@@ -107,24 +107,24 @@ weave_child(WEAVE_ARGS[:benchmarkbody], mod = @__MODULE__, args = WEAVE_ARGS)
 end
 
 model_def = demo4
-data = (100_000, );
+data = (100_000,);
 ```
 
 ```julia; echo=false
-weave_child(WEAVE_ARGS[:benchmarkbody], mod = @__MODULE__, args = WEAVE_ARGS)
+weave_child(WEAVE_ARGS[:benchmarkbody]; mod=@__MODULE__, args=WEAVE_ARGS)
 ```
 
 ```julia
 @model function demo4_dotted(n, ::Type{TV}=Vector{Float64}) where {TV}
     m ~ Normal()
     x = TV(undef, n)
-    x .~ Normal(m, 1.0)
+    return x .~ Normal(m, 1.0)
 end
 
 model_def = demo4_dotted
-data = (100_000, );
+data = (100_000,);
 ```
 
 ```julia; echo=false
-weave_child(WEAVE_ARGS[:benchmarkbody], mod = @__MODULE__, args = WEAVE_ARGS)
+weave_child(WEAVE_ARGS[:benchmarkbody]; mod=@__MODULE__, args=WEAVE_ARGS)
 ```

docs/src/index.md

@@ -1,4 +1,3 @@
 # DynamicPPL.jl
 
 A domain-specific language and backend for probabilistic programming languages, used by [Turing.jl](https://github.com/TuringLang/Turing.jl).
-

docs/src/tutorials/prob-interface.md

@@ -47,6 +47,7 @@ nothing # hide
 ```
 
 We can also decondition only some of the variables:
+
 ```@example probinterface
 decondition(model, :μ)
 nothing # hide

src/submodel_macro.jl

@@ -120,49 +120,49 @@ julia> @model inner() = x ~ Normal()
 inner (generic function with 2 methods)
 
 julia> # When `prefix` is unspecified, no prefix is used.
-       @model outer() = @submodel a = inner()
-outer (generic function with 2 methods)
+       @model submodel_noprefix() = @submodel a = inner()
+submodel_noprefix (generic function with 2 methods)
 
-julia> @varname(x) in keys(VarInfo(outer()))
+julia> @varname(x) in keys(VarInfo(submodel_noprefix()))
 true
 
 julia> # Explicitely don't use any prefix.
-       @model outer() = @submodel prefix=false a = inner()
-outer (generic function with 2 methods)
+       @model submodel_prefix_false() = @submodel prefix=false a = inner()
+submodel_prefix_false (generic function with 2 methods)
 
-julia> @varname(x) in keys(VarInfo(outer()))
+julia> @varname(x) in keys(VarInfo(submodel_prefix_false()))
 true
 
 julia> # Automatically determined from `a`.
-       @model outer() = @submodel prefix=true a = inner()
-outer (generic function with 2 methods)
+       @model submodel_prefix_true() = @submodel prefix=true a = inner()
+submodel_prefix_true (generic function with 2 methods)
 
-julia> @varname(var"a.x") in keys(VarInfo(outer()))
+julia> @varname(var"a.x") in keys(VarInfo(submodel_prefix_true()))
 true
 
 julia> # Using a static string.
-       @model outer() = @submodel prefix="my prefix" a = inner()
-outer (generic function with 2 methods)
+       @model submodel_prefix_string() = @submodel prefix="my prefix" a = inner()
+submodel_prefix_string (generic function with 2 methods)
 
-julia> @varname(var"my prefix.x") in keys(VarInfo(outer()))
+julia> @varname(var"my prefix.x") in keys(VarInfo(submodel_prefix_string()))
 true
 
 julia> # Using string interpolation.
-       @model outer() = @submodel prefix="\$(nameof(inner()))" a = inner()
-outer (generic function with 2 methods)
+       @model submodel_prefix_interpolation() = @submodel prefix="\$(nameof(inner()))" a = inner()
+submodel_prefix_interpolation (generic function with 2 methods)
 
-julia> @varname(var"inner.x") in keys(VarInfo(outer()))
+julia> @varname(var"inner.x") in keys(VarInfo(submodel_prefix_interpolation()))
 true
 
 julia> # Or using some arbitrary expression.
-       @model outer() = @submodel prefix=1 + 2 a = inner()
-outer (generic function with 2 methods)
+       @model submodel_prefix_expr() = @submodel prefix=1 + 2 a = inner()
+submodel_prefix_expr (generic function with 2 methods)
 
-julia> @varname(var"3.x") in keys(VarInfo(outer()))
+julia> @varname(var"3.x") in keys(VarInfo(submodel_prefix_expr()))
 true
 
 julia> # (×) Automatic prefixing without a left-hand side expression does not work!
-       @model outer() = @submodel prefix=true inner()
+       @model submodel_prefix_error() = @submodel prefix=true inner()
 ERROR: LoadError: cannot automatically prefix with no left-hand side
 [...]
 ```

test/lkj.jl

@@ -1,7 +1,13 @@
-using Bijectors: pd_from_upper, pd_from_lower
-
 function pd_from_triangular(X::AbstractMatrix, uplo::Char)
-    return uplo == 'U' ? pd_from_upper(X) : pd_from_lower(X)
+    # Pre-allocation fixes a problem with abstract element types in Julia 1.10
+    # Ref https://github.com/TuringLang/DynamicPPL.jl/pull/570#issue-2092729916
+    out = similar(X, Base.promote_op(*, eltype(X), eltype(X)))
+    if uplo === 'U'
+        mul!(out, UpperTriangular(X)', UpperTriangular(X))
+    else
+        mul!(out, LowerTriangular(X), LowerTriangular(X)')
+    end
+    return out
 end
 
 @model lkj_prior_demo() = x ~ LKJ(2, 1)

test/turing/Project.toml

@@ -6,6 +6,6 @@ Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
 Turing = "fce5fe82-541a-59a6-adf8-730c64b5f9a0"
 
 [compat]
-DynamicPPL = "0.20, 0.21, 0.22, 0.23, 0.24"
-Turing = "0.21 - 0.30"
-julia = "1.6"
+DynamicPPL = "0.24"
+Turing = "0.30"
+julia = "1.7"

test/turing/compiler.jl

@@ -110,7 +110,9 @@
         end
 
         chain = sample(
-            newinterface(obs), HMC{Turing.ForwardDiffAD{2}}(0.75, 3, :p, :x), 100
+            newinterface(obs),
+            HMC(0.75, 3, :p, :x; adtype=AutoForwardDiff(; chunksize=2)),
+            100,
         )
     end
     @testset "no return" begin
@@ -186,8 +188,7 @@
 
         # Vector assumptions
         N = 10
-        setchunksize(N)
-        alg = HMC(0.2, 4)
+        alg = HMC(0.2, 4; adtype=AutoForwardDiff(; chunksize=N))
 
         @model function vdemo3()
             x = Vector{Real}(undef, N)
@@ -254,8 +255,7 @@
 
         # Vector assumptions
         N = 10
-        setchunksize(N)
-        alg = HMC(0.2, 4)
+        alg = HMC(0.2, 4; adtype=AutoForwardDiff(; chunksize=N))
 
         @model function vdemo3()
             x = Vector{Real}(undef, N)
@@ -300,8 +300,7 @@
     end
     @testset "Type parameters" begin
         N = 10
-        setchunksize(N)
-        alg = HMC(0.01, 5)
+        alg = HMC(0.01, 5; adtype=AutoForwardDiff(; chunksize=N))
         x = randn(1000)
         @model function vdemo1(::Type{T}=Float64) where {T}
             x = Vector{T}(undef, N)