Make the non-essential stuff all keyword arguments

Author: penelopeysm

ext/DynamicPPLMarginalLogDensitiesExt.jl

@@ -9,7 +9,7 @@ _to_varname(n::VarName) = n
 """
     marginalize(
         model::DynamicPPL.Model,
-        varnames::AbstractVector{<:Union{Symbol,<:VarName}},
+        varnames::AbstractVector{<:Union{Symbol,<:VarName}};
         varinfo::DynamicPPL.AbstractVarInfo=link(VarInfo(model), model),
         getlogprob=DynamicPPL.getlogjoint,
         method::MarginalLogDensities.AbstractMarginalizer=MarginalLogDensities.LaplaceApprox();
@@ -23,17 +23,22 @@ log-density of the given `model`, after marginalizing out the variables specifie
 The resulting object can be called with a vector of parameter values to compute the marginal
 log-density.
 
-You can specify the `varinfo` to use for the model. By default we use a linked `VarInfo`,
-meaning that the resulting log-density function accepts parameters that have been
-transformed to unconstrained space.
+## Keyword arguments
 
-The `getlogprob` argument can be used to specify which kind of marginal log-density to
-compute. Its default value is `DynamicPPL.getlogjoint` which returns the marginal log-joint
-probability.
+- `varinfo`: The `varinfo` to use for the model. By default we use a linked `VarInfo`,
+   meaning that the resulting log-density function accepts parameters that have bee_FWDn
+   transformed to unconstrained space.
 
-By default the marginalization is performed with a Laplace approximation. Please see [the
-MarginalLogDensities.jl package](https://github.com/ElOceanografo/MarginalLogDensities.jl/)
-for other options.
+- `getlogprob`: A function which specifies which kind of marginal log-density to compute.
+   Its default value is `DynamicPPL.getlogjoint` which returns the marginal log-joint
+   probability.
+
+- `method`: The marginalization method; defaults to a Laplace approximation. Please see [the
+   MarginalLogDensities.jl package](https://github.com/ElOceanografo/MarginalLogDensities.jl/)
+   for other options.
+
+- Other keyword arguments are passed to the `MarginalLogDensities.MarginalLogDensity`
+  constructor.
 
 ## Example
 
@@ -58,10 +63,10 @@ julia> logpdf(Normal(2.0), 1.0)
 """
 function DynamicPPL.marginalize(
     model::DynamicPPL.Model,
-    varnames::AbstractVector{<:Union{Symbol,<:VarName}},
+    varnames::AbstractVector{<:Union{Symbol,<:VarName}};
     varinfo::DynamicPPL.AbstractVarInfo=DynamicPPL.link(DynamicPPL.VarInfo(model), model),
     getlogprob::Function=DynamicPPL.getlogjoint,
-    method::MarginalLogDensities.AbstractMarginalizer=MarginalLogDensities.LaplaceApprox();
+    method::MarginalLogDensities.AbstractMarginalizer=MarginalLogDensities.LaplaceApprox(),
     kwargs...,
 )
     # Determine the indices for the variables to marginalise out.

test/ext/DynamicPPLMarginalLogDensitiesExt.jl

@@ -17,7 +17,11 @@ using ADTypes: AutoForwardDiff
         for vn in [@varname(x), :x]
             for getlogprob in [DynamicPPL.getlogprior, DynamicPPL.getlogjoint]
                 marginalized = marginalize(
-                    model, [vn], vi, getlogprob; hess_adtype=AutoForwardDiff()
+                    model,
+                    [vn];
+                    varinfo=vi,
+                    getlogprob=getlogprob,
+                    hess_adtype=AutoForwardDiff(),
                 )
                 for y in range(-5, 5; length=100)
                     @test marginalized([y]) ≈ logpdf(Normal(0, 1), y) atol = 1e-5
@@ -36,27 +40,29 @@ using ADTypes: AutoForwardDiff
         vi_linked = DynamicPPL.link(vi_unlinked, model)
 
         @testset "unlinked VarInfo" begin
-            mx = marginalize(model, [@varname(x)], vi_unlinked)
+            mx = marginalize(model, [@varname(x)]; varinfo=vi_unlinked)
             for x in range(0.01, 0.99; length=10)
                 @test mx([x]) ≈ logpdf(Beta(2, 2), x)
             end
             # generally when marginalising Beta it doesn't go to zero
-            my = marginalize(model, [@varname(y)], vi_unlinked)
+            # https://github.com/TuringLang/DynamicPPL.jl/pull/1036#discussion_r2349388067
+            my = marginalize(model, [@varname(y)]; varinfo=vi_unlinked)
             diff = my([0.0]) - logpdf(Normal(), 0.0)
             for x in range(-5, 5; length=10)
                 @test my([x]) ≈ logpdf(Normal(), x) + diff
             end
         end
 
         @testset "linked VarInfo" begin
-            mx = marginalize(model, [@varname(x)], vi_linked)
+            mx = marginalize(model, [@varname(x)]; varinfo=vi_linked)
             binv = Bijectors.inverse(Bijectors.bijector(Beta(2, 2)))
             for y_linked in range(-5, 5; length=10)
                 y_unlinked = binv(y_linked)
                 @test mx([y_linked]) ≈ logpdf(Beta(2, 2), y_unlinked)
             end
             # generally when marginalising Beta it doesn't go to zero
-            my = marginalize(model, [@varname(y)], vi_linked)
+            # https://github.com/TuringLang/DynamicPPL.jl/pull/1036#discussion_r2349388067
+            my = marginalize(model, [@varname(y)]; varinfo=vi_linked)
             diff = my([0.0]) - logpdf(Normal(), 0.0)
             for x in range(-5, 5; length=10)
                 @test my([x]) ≈ logpdf(Normal(), x) + diff