Merge pull request #578 from SciML/docsup

Fix mtk empty contraints creation and update rosenbrock example

Project.toml

@@ -1,6 +1,6 @@
 name = "Optimization"
 uuid = "7f7a1694-90dd-40f0-9382-eb1efda571ba"
-version = "3.16.0"
+version = "3.17.0"
 
 [deps]
 ADTypes = "47edcb42-4c32-4615-8424-f2b9edc5f35b"

docs/src/examples/rosenbrock.md

@@ -79,13 +79,22 @@ optf = OptimizationFunction(rosenbrock, Optimization.AutoForwardDiff(); cons = c
 prob = OptimizationProblem(optf, x0, _p, lcons = [-Inf], ucons = [0.25^2])
 sol = solve(prob, IPNewton()) # -Inf < cons_circ(sol.u, _p) = 0.25^2
 
+## Evolutionary.jl Solvers
+
+using OptimizationEvolutionary
+sol = solve(prob, CMAES(μ = 40, λ = 100), abstol = 1e-15) # -Inf < cons_circ(sol.u, _p) = 0.25^2
+
+## IPOPT through OptimizationMOI
+
+using OptimizationMOI, Ipopt
+
 function con2_c(res, x, p)
     res .= [x[1]^2 + x[2]^2, x[2] * sin(x[1]) - x[1]]
 end
 
-optf = OptimizationFunction(rosenbrock, Optimization.AutoForwardDiff(); cons = con2_c)
+optf = OptimizationFunction(rosenbrock, Optimization.AutoZygote(); cons = con2_c)
 prob = OptimizationProblem(optf, x0, _p, lcons = [-Inf, -Inf], ucons = [Inf, Inf])
-sol = solve(prob, IPNewton())
+sol = solve(prob, Ipopt.Optimizer())
 
 # Now let's switch over to OptimizationOptimisers with reverse-mode AD
 
@@ -112,12 +121,7 @@ sol = solve(prob, Opt(:LD_LBFGS, 2))
 
 prob = OptimizationProblem(optf, x0, _p, lb = [-1.0, -1.0], ub = [0.8, 0.8])
 sol = solve(prob, Opt(:LD_LBFGS, 2))
-# sol = solve(prob, Opt(:G_MLSL_LDS, 2), nstart=2, local_method = Opt(:LD_LBFGS, 2), maxiters=10000)
-
-## Evolutionary.jl Solvers
-
-using OptimizationEvolutionary
-sol = solve(prob, CMAES(μ = 40, λ = 100), abstol = 1e-15) # -1.0 ≤ x[1], x[2] ≤ 0.8
+sol = solve(prob, Opt(:G_MLSL_LDS, 2), local_method = Opt(:LD_LBFGS, 2), maxiters = 10000) #a global optimizer with random starts of local optimization
 
 ## BlackBoxOptim.jl Solvers
 

ext/OptimizationMTKExt.jl

@@ -31,11 +31,16 @@ function Optimization.instantiate_function(f, x, adtype::AutoModelingToolkit, p,
         H .= res * v
     end
 
-    cons = (res, θ) -> f.cons(res, θ, p)
-
-    cons_j = (J, θ) -> f.cons_j(J, θ, p)
+    if !isnothing(f.cons)
+        cons = (res, θ) -> f.cons(res, θ, p)
+        cons_j = (J, θ) -> f.cons_j(J, θ, p)
+        cons_h = (res, θ) -> f.cons_h(res, θ, p)
+    else
+        cons = nothing
+        cons_j = nothing
+        cons_h = nothing
+    end
 
-    cons_h = (res, θ) -> f.cons_h(res, θ, p)
     return OptimizationFunction{true}(f.f, adtype; grad = grad, hess = hess, hv = hv,
         cons = cons, cons_j = cons_j, cons_h = cons_h,
         hess_prototype = f.hess_prototype,
@@ -71,11 +76,16 @@ function Optimization.instantiate_function(f, cache::Optimization.ReInitCache,
         H .= res * v
     end
 
-    cons = (res, θ) -> f.cons(res, θ, cache.p)
-
-    cons_j = (J, θ) -> f.cons_j(J, θ, cache.p)
+    if !isnothing(f.cons)
+        cons = (res, θ) -> f.cons(res, θ, cache.p)
+        cons_j = (J, θ) -> f.cons_j(J, θ, cache.p)
+        cons_h = (res, θ) -> f.cons_h(res, θ, cache.p)
+    else
+        cons = nothing
+        cons_j = nothing
+        cons_h = nothing
+    end
 
-    cons_h = (res, θ) -> f.cons_h(res, θ, cache.p)
     return OptimizationFunction{true}(f.f, adtype; grad = grad, hess = hess, hv = hv,
         cons = cons, cons_j = cons_j, cons_h = cons_h,
         hess_prototype = f.hess_prototype,