feat(breaking): update to Manopt 0.5

.github/workflows/CI.yml

@@ -23,7 +23,7 @@ jobs:
       fail-fast: false
       matrix:
         version:
-          - '1.10'
+          - '1.11'
         os:
           - ubuntu-latest
         arch:

Project.toml

@@ -26,10 +26,10 @@ StaticTools = "86c06d3c-3f03-46de-9781-57580aa96d0a"
 StatsFuns = "4c63d2b9-4356-54db-8cca-17b64c39e42c"
 
 [compat]
-BayesBase = "1.3"
+BayesBase = "1.5.0"
 Bumper = "0.6"
 Distributions = "0.25"
-ExponentialFamily = "1.5"
+ExponentialFamily = "1.6"
 ExponentialFamilyManifolds = "1.5"
 FastCholesky = "1.3"
 FillArrays = "1"
@@ -38,7 +38,7 @@ LinearAlgebra = "1.10"
 LoopVectorization = "0.12"
 Manifolds = "0.9"
 ManifoldsBase = "0.15"
-Manopt = "0.4"
+Manopt = "0.5"
 Random = "1.10"
 RecursiveArrayTools = "3.2"
 StableRNGs = "1"

src/manopt/bounded_norm_update_rule.jl

@@ -3,7 +3,7 @@ import Manopt
 using LoopVectorization
 
 """
-    BoundedNormUpdateRule(limit; direction = IdentityUpdateRule()) 
+    BoundedNormUpdateRule(limit; direction = Manopt.IdentityUpdateRule()) 
 
 A `DirectionUpdateRule` is a direction rule that constrains the norm of the direction to a specified limit.
 
@@ -23,7 +23,7 @@ struct BoundedNormUpdateRule{L,D} <: Manopt.DirectionUpdateRule
     direction::D
 end
 
-function BoundedNormUpdateRule(limit; direction = IdentityUpdateRule())
+function BoundedNormUpdateRule(limit; direction = Manopt.IdentityUpdateRule())
     return BoundedNormUpdateRule(limit, direction)
 end
 

src/projected_to.jl

@@ -108,7 +108,7 @@ The following parameters are available:
 * `strategy = ExponentialFamilyProjection.DefaultStrategy()`: The strategy to use to compute the gradients.
 * `niterations = 100`: The number of iterations for the optimization procedure.
 * `tolerance = 1e-6`: The tolerance for the norm of the gradient.
-* `stepsize = ConstantStepsize(0.1)`: The stepsize for the optimization procedure. Accepts stepsizes from `Manopt.jl`.
+* `stepsize = ConstantLength(0.1)`: The stepsize for the optimization procedure. Accepts stepsizes from `Manopt.jl`.
 * `seed`: Optional; Seed for the `rng`
 * `rng`: Optional; Random number generator
 * `direction = BoundedNormUpdateRule(static(1.0)`: Direction update rule. Accepts `Manopt.DirectionUpdateRule` from `Manopt.jl`.
@@ -117,7 +117,7 @@ Base.@kwdef struct ProjectionParameters{S,I,T,P,D,N,U}
     strategy::S = DefaultStrategy()
     niterations::I = 100
     tolerance::T = 1e-6
-    stepsize::P = ConstantStepsize(0.1)
+    stepsize::P = ConstantLength(0.1)
     seed::D = 42
     rng::N = StableRNG(seed)
     direction::U = BoundedNormUpdateRule(static(1.0))

test/manopt/bounded_norm_update_rule_tests.jl

@@ -13,10 +13,10 @@
     # returning the unbounded gradient as the first argument and a collection of bounded gradients for testing.
     function apply_update_rules_for_test(p, limit)
         cpa = DefaultManoptProblem(M, ManifoldGradientObjective(f, grad_f))
-        gst = GradientDescentState(M, zero(p))
+        gst = GradientDescentState(M; p=zero(p))
         Manopt.set_iterate!(gst, M, p)
 
-        _, X = IdentityUpdateRule()(cpa, gst, 1)
+        _, X = Manopt.IdentityUpdateRule()(cpa, gst, 1)
         X_identity = copy(X)
 
         _, X = BoundedNormUpdateRule(limit)(cpa, gst, 1)
@@ -79,7 +79,7 @@
     @testset "JET tests" begin
         for limit in (1, 1.0, 1.0f0), p in (zeros(Float64, 3), zeros(Float32, 3))
             cpa = DefaultManoptProblem(M, ManifoldGradientObjective(f, grad_f))
-            gst = GradientDescentState(M, zero(p))
+            gst = GradientDescentState(M; p=zero(p))
             @test_opt BoundedNormUpdateRule(limit)(cpa, gst, 1)
             @test_opt BoundedNormUpdateRule(static(limit))(cpa, gst, 1)
             @test_opt BoundedNormUpdateRule(

test/projection/projected_to_setuptests.jl

@@ -93,7 +93,7 @@ function test_convergence_nsamples(
     nsamples_tolerance = _convergence_nsamples_default_tolerance(distribution),
     nsamples_niterations = _convergence_nsamples_default_niterations(distribution),
     nsamples_rng = StableRNG(42),
-    nsamples_stepsize = ConstantStepsize(0.1),
+    nsamples_stepsize = ConstantLength(0.1),
     nsamples_required_accuracy = 1e-1,
     kwargs...,
 )
@@ -168,7 +168,7 @@ function test_convergence_niterations(
     niterations_tolerance = _convergence_niterations_default_tolerance(distribution),
     niterations_nsamples = _convergence_niterations_default_nsamples(distribution),
     niterations_rng = StableRNG(42),
-    niterations_stepsize = ConstantStepsize(0.1),
+    niterations_stepsize = ConstantLength(0.1),
     niterations_required_accuracy = 1e-1,
     kwargs...,
 )
@@ -219,7 +219,7 @@ function test_convergence_niterations_mle(
     niterations_tolerance = _convergence_niterations_default_tolerance(distribution),
     niterations_nsamples = _convergence_niterations_default_nsamples(distribution),
     niterations_rng = StableRNG(42),
-    niterations_stepsize = ConstantStepsize(0.1),
+    niterations_stepsize = ConstantLength(0.1),
     niterations_required_accuracy = 1e-1,
     kwargs...,
 )
@@ -268,7 +268,7 @@ function test_convergence_nsamples_mle(
     nsamples_tolerance = _convergence_nsamples_default_tolerance(distribution),
     nsamples_niterations = _convergence_nsamples_default_niterations(distribution),
     nsamples_rng = StableRNG(42),
-    nsamples_stepsize = ConstantStepsize(0.1),
+    nsamples_stepsize = ConstantLength(0.1),
     nsamples_required_accuracy = 1e-1,
     kwargs...,
 )

test/projection/projected_to_tests.jl

@@ -54,7 +54,7 @@
     @test typeof(get_stopping_criterion(defaultparams)) ==
           typeof(get_stopping_criterion(parameters_from_creation))
     # These should pass as soon as `Manopt` implements `==`
-    @test_broken getstepsize(defaultparams) == getstepsize(parameters_from_creation)
+    @test getstepsize(defaultparams) == getstepsize(parameters_from_creation)
     @test_broken get_stopping_criterion(defaultparams) ==
                  get_stopping_criterion(parameters_from_creation)
 
@@ -518,7 +518,7 @@ end
         nothing,
     )
     initialpoint = rand(manifold)
-    direction = MomentumGradient(manifold, initialpoint)
+    direction = MomentumGradient(p=initialpoint)
 
     momentum_parameters =
         ProjectionParameters(direction = direction, niterations = 1000, tolerance = 1e-8)
@@ -529,7 +529,7 @@ end
 
     @test approximated isa MvNormalMeanCovariance
     @test kldivergence(approximated, true_dist) < 0.01
-    @test projection.parameters.direction isa MomentumGradient
+    @test projection.parameters.direction isa Manopt.ManifoldDefaultsFactory
 end
 
 @testitem "MomentumGradient direction update rule on samples" begin
@@ -547,7 +547,7 @@ end
     )
 
     initialpoint = rand(rng, manifold)
-    direction = MomentumGradient(manifold, initialpoint)
+    direction = MomentumGradient(p=initialpoint)
 
     momentum_parameters =
         ProjectionParameters(direction = direction, niterations = 1000, tolerance = 1e-8)
@@ -557,5 +557,5 @@ end
 
     @test approximated isa MvNormalMeanCovariance
     @test kldivergence(approximated, true_dist) < 0.01  # Ensure good approximation
-    @test projection.parameters.direction isa MomentumGradient
+    @test projection.parameters.direction isa Manopt.ManifoldDefaultsFactory
 end