simplify test info

Author: xukai92

test/hmc.jl

@@ -10,15 +10,15 @@ n_adapts = 1_000
 
 @testset "HMC and NUTS" begin
     temp = randn(D,100)
-    @testset "$metric" for metric in [UnitEuclideanMetric(θ_init),
+    @testset "$typeof(metric)" for metric in [UnitEuclideanMetric(θ_init),
                                       DiagEuclideanMetric(θ_init, vec(var(temp; dims=2))),
                                       DenseEuclideanMetric(θ_init, cov(temp'))]
         h = Hamiltonian(metric, logπ, ∂logπ∂θ)
-        @testset "$prop" for prop in [TakeLastProposal(Leapfrog(ϵ), n_steps),
+        @testset "$typeof(prop)" for prop in [TakeLastProposal(Leapfrog(ϵ), n_steps),
                                       SliceNUTS(Leapfrog(find_good_eps(h, θ_init)))]
             samples = AdvancedHMC.sample(h, prop, θ_init, n_samples; verbose=false)
             @test mean(samples) ≈ zeros(D) atol=RNDATOL
-            @testset "$adapter" for adapter in [PreConditioner(metric), DualAveraging(0.8, prop.integrator.ϵ),
+            @testset "$typeof(adapter)" for adapter in [PreConditioner(metric), DualAveraging(0.8, prop.integrator.ϵ),
                                                 NaiveCompAdapter(PreConditioner(metric), DualAveraging(0.8, prop.integrator.ϵ)),
                                                 ThreePhaseAdapter(n_adapts, PreConditioner(metric), DualAveraging(0.8, prop.integrator.ϵ))]
                 samples = AdvancedHMC.sample(h, prop, θ_init, n_samples, adapter, n_adapts; verbose=false)