fix: update ExponetialFamily.jl

ReactiveBayes · Jul 25, 2024 · 99b234b · 99b234b
1 parent 68e2f0d
commit 99b234b
Show file tree

Hide file tree

Showing 3 changed files with 5 additions and 5 deletions.
diff --git a/Project.toml b/Project.toml
@@ -15,7 +15,7 @@ Static = "aedffcd0-7271-4cad-89d0-dc628f76c6d3"
 
 [compat]
 BayesBase = "1.3"
-ExponentialFamily = "1.4.3"
+ExponentialFamily = "1.5.1"
 LinearAlgebra = "1.10"
 Manifolds = "0.9"
 ManifoldsBase = "0.15"

diff --git a/src/single_point_manifold.jl b/src/single_point_manifold.jl
@@ -26,7 +26,7 @@ ManifoldsBase.injectivity_radius(M::SinglePointManifold) = zero(eltype(M.point))
 ManifoldsBase.default_retraction_method(::SinglePointManifold) = ExponentialRetraction()
 
 function ManifoldsBase.check_point(M::SinglePointManifold, p; kwargs...)
-    if p[1] != M.point
+    if !(p ≈ M.point)
         return DomainError(p, "The point $(p) does not lie on $(M), which contains only $(M.point).")
     end
     return nothing

diff --git a/test/natural_manifolds/categorical_tests.jl b/test/natural_manifolds/categorical_tests.jl
@@ -26,12 +26,12 @@ end
 
     function f(M, p)
         ef = convert(ExponentialFamilyDistribution, M, p)
-        return (mean(ef) - 0.5)^2
+        η = getnaturalparameters(ef)
+        return (mean(η) - 0.5)^2
     end
 
     function g(M, p)
-        ef = convert(ExponentialFamilyDistribution, M, p)
-        return project(M, p, 2 * (mean(ef) - 0.5) * p ./ 10)
+        return project(M, p, 2 * p ./ 10)
     end
 
     q = gradient_descent(M, f, g, rand(rng, M))