add unconstrained_api

JuliaSmoothOptimizers · Jul 26, 2021 · 02d5edd · 02d5edd
1 parent d728aa5
commit 02d5edd
Show file tree

Hide file tree

Showing 5 changed files with 127 additions and 0 deletions.
diff --git a/src/nlp/consistency.jl b/src/nlp/consistency.jl
@@ -23,6 +23,7 @@ function consistent_nlps(
 )
   consistent_counters(nlps)
   test_meta && consistent_meta(nlps, rtol = rtol)
+  unconstrained_api(nlps, exclude = exclude)
   consistent_functions(nlps, rtol = rtol, exclude = exclude)
   consistent_counters(nlps)
   for nlp in nlps
@@ -59,6 +60,23 @@ function consistent_nlps(
   end
 end
 
+function unconstrained_api(nlps; exclude = [])
+  for nlp in nlps
+    if nlp.meta.ncon == 0
+      x0, x1 = nlp.meta.x0, ones(nlp.meta.nvar)
+      @test nlp.meta.nnzj == 0
+      @test cons in exclude || cons(nlp, x0) == []
+      @test jac_coord in exclude || jac_coord(nlp, x0) == []
+      @test jac_coord in exclude || jac_structure(nlp) == ([], [])
+      @test jac in exclude || jac(nlp, x0) == zeros(0, nlp.meta.nvar)
+      @test jtprod in exclude || all(jtprod(nlp, x0, nlp.meta.y0) .== 0)
+      @test jprod in exclude || jprod(nlp, x0, x1) == []
+      @test hess in exclude || hess(nlp, x0, ones(0)) == hess(nlp, x0)
+      @test hprod in exclude || hprod(nlp, x0, ones(0), x1) == hprod(nlp, x0, x1)
+    end
+  end
+end
+
 function consistent_meta(nlps; rtol = 1.0e-8)
   fields = [:nvar, :x0, :lvar, :uvar, :ifix, :ilow, :iupp, :irng, :ifree, :ncon, :y0]
   N = length(nlps)

diff --git a/src/nlp/problems/brownden.jl b/src/nlp/problems/brownden.jl
@@ -129,3 +129,39 @@ function NLPModels.hprod!(
   end
   return Hv
 end
+
+function NLPModels.cons!(nlp::BROWNDEN, x::AbstractVector{T}, c) where {T}
+  @lencheck nlp.meta.ncon c
+  increment!(nlp, :neval_cons)
+  return c
+end
+function NLPModels.jac_structure!(
+  nlp::BROWNDEN,
+  rows::AbstractVector{T},
+  cols::AbstractVector{T},
+) where {T}
+  @lencheck nlp.meta.nnzj rows cols
+  return rows, cols
+end
+function NLPModels.jac_coord!(nlp::BROWNDEN, x::AbstractVector{T}, vals) where {T}
+  @lencheck nlp.meta.nnzj vals
+  increment!(nlp, :neval_jac)
+  return vals
+end
+function NLPModels.jprod!(nlp::BROWNDEN, x::AbstractVector{T}, v, Jv) where {T}
+  @lencheck nlp.meta.nvar v
+  @lencheck nlp.meta.ncon Jv
+  increment!(nlp, :neval_jprod)
+  return Jv
+end
+function NLPModels.jtprod!(nlp::BROWNDEN, x::AbstractVector{T}, v, Jtv) where {T}
+  @lencheck nlp.meta.nvar Jtv
+  @lencheck nlp.meta.ncon v
+  increment!(nlp, :neval_jtprod)
+  fill!(Jtv, zero(T))
+  return Jtv
+end
+function NLPModels.hess_coord!(nlp::BROWNDEN, x, y, vals; obj_weight=1.0)
+  return hess_coord!(nlp, x, vals; obj_weight=obj_weight)
+end
+NLPModels.hprod!(nlp::BROWNDEN, x, y, v, Hv; obj_weight=1.0) = hprod!(nlp, x, v, Hv; obj_weight=obj_weight)
diff --git a/src/nlp/problems/hs5.jl b/src/nlp/problems/hs5.jl
@@ -76,3 +76,39 @@ function NLPModels.hprod!(
     (-sin(x[1] + x[2]) * (v[1] + v[2]) * ones(T, 2) + 2 * [v[1] - v[2]; v[2] - v[1]]) * obj_weight
   return Hv
 end
+
+function NLPModels.cons!(nlp::HS5, x::AbstractVector{T}, c) where {T}
+  @lencheck nlp.meta.ncon c
+  increment!(nlp, :neval_cons)
+  return c
+end
+function NLPModels.jac_structure!(
+  nlp::HS5,
+  rows::AbstractVector{T},
+  cols::AbstractVector{T},
+) where {T}
+  @lencheck nlp.meta.nnzj rows cols
+  return rows, cols
+end
+function NLPModels.jac_coord!(nlp::HS5, x::AbstractVector{T}, vals) where {T}
+  @lencheck nlp.meta.nnzj vals
+  increment!(nlp, :neval_jac)
+  return vals
+end
+function NLPModels.jprod!(nlp::HS5, x::AbstractVector{T}, v, Jv) where {T}
+  @lencheck nlp.meta.nvar v
+  @lencheck nlp.meta.ncon Jv
+  increment!(nlp, :neval_jprod)
+  return Jv
+end
+function NLPModels.jtprod!(nlp::HS5, x::AbstractVector{T}, v, Jtv) where {T}
+  @lencheck nlp.meta.nvar Jtv
+  @lencheck nlp.meta.ncon v
+  increment!(nlp, :neval_jtprod)
+  fill!(Jtv, zero(T))
+  return Jtv
+end
+function NLPModels.hess_coord!(nlp::HS5, x, y, vals; obj_weight=1.0)
+  return hess_coord!(nlp, x, vals; obj_weight=obj_weight)
+end
+NLPModels.hprod!(nlp::HS5, x, y, v, Hv; obj_weight=1.0) = hprod!(nlp, x, v, Hv; obj_weight=obj_weight)
diff --git a/src/nls/consistency.jl b/src/nls/consistency.jl
@@ -29,6 +29,7 @@ function consistent_nlss(
     reset!(nls)
   end
   consistent_functions(nlss, exclude = exclude)
+  unconstrained_api(nlss, exclude = exclude)
 
   if test_slack && has_inequalities(nlss[1])
     reset!.(nlss)

diff --git a/src/nls/problems/mgh01.jl b/src/nls/problems/mgh01.jl
@@ -179,3 +179,39 @@ function NLPModels.hprod!(
   Hv .= obj_weight * [T(1) - 200 * x[2]+600 * x[1]^2 -200*x[1]; -200*x[1] T(100)] * v
   return Hv
 end
+
+function NLPModels.cons!(nls::MGH01, x::AbstractVector{T}, c) where {T}
+  @lencheck nls.meta.ncon c
+  increment!(nls, :neval_cons)
+  return c
+end
+function NLPModels.jac_structure!(
+  nls::MGH01,
+  rows::AbstractVector{T},
+  cols::AbstractVector{T},
+) where {T}
+  @lencheck nls.meta.nnzj rows cols
+  return rows, cols
+end
+function NLPModels.jac_coord!(nls::MGH01, x::AbstractVector{T}, vals) where {T}
+  @lencheck nls.meta.nnzj vals
+  increment!(nls, :neval_jac)
+  return vals
+end
+function NLPModels.jprod!(nls::MGH01, x::AbstractVector{T}, v, Jv) where {T}
+  @lencheck nls.meta.nvar v
+  @lencheck nls.meta.ncon Jv
+  increment!(nls, :neval_jprod)
+  return Jv
+end
+function NLPModels.jtprod!(nls::MGH01, x::AbstractVector{T}, v, Jtv) where {T}
+  @lencheck nls.meta.nvar Jtv
+  @lencheck nls.meta.ncon v
+  increment!(nls, :neval_jtprod)
+  fill!(Jtv, zero(T))
+  return Jtv
+end
+function NLPModels.hess_coord!(nls::MGH01, x, y, vals; obj_weight=1.0)
+  return hess_coord!(nls, x, vals; obj_weight=obj_weight)
+end
+NLPModels.hprod!(nls::MGH01, x, y, v, Hv; obj_weight=1.0) = hprod!(nls, x, v, Hv; obj_weight=obj_weight)