JuMP - MethodError: `getindex` has no method matching getindex(::Function, ::Int64)

Question

I have to calculate the ideal vector of this multiobjective problem. I cannot how to access the first and second function of the functs_BK1() in the @objective. Any Idea how I could make all dynamic and support n-functions?

using JuMP
using Ipopt


function functs_BK1(x::Vector)
    f = zeros(2)
    f[1] = x[1]^2 + x[2]^2
    f[2] = (x[1]-5.0)^2 + (x[2]-5.0)^2

    return f
end

function bounds_BK1()
    return ([-5.0;-5.0],[10.0;10.0])
end


m = Model(solver=IpoptSolver(print_level=0))
@variable(m, x[i=1:2])

@NLexpression(m, F1, functs_BK1()[1]) #<--- Problem is here
@NLexpression(m, F2, functs_BK1()[2]) #<--- here

@constraint(m, x .>= bounds_BK1()[1])
@constraint(m, x .<= bounds_BK1()[2])
s=solve(m)

score 1 · Answer 1 · answered Aug 30 '16 at 15:50

In your function functs_BK1, you have defined f to be an array of Float64 values, while the @NLexpression macro must build up the desired expression from JuMP variables as its third argument. As stated in the JuMP documentation for Nonlinear Expressions, all nonlinear expressions must be defined inside of the @NLexpression macro, and AffExpr and QuadExpr objects cannot currently be used inside of the @NLexpression macro.

The following set of commands results in finding a solution via the solve operation:

julia> using JuMP

julia> using Ipopt

julia> m = Model(solver=IpoptSolver(print_level=0))
Feasibility problem with:
 * 0 linear constraints
 * 0 variables
Solver is Ipopt

julia> @variable(m, x[i=1:2])
2-element Array{JuMP.Variable,1}:
 x[1]
 x[2]

julia> function bounds_BK1()
           return ([-5.0;-5.0],[10.0;10.0])
       end
bounds_BK1 (generic function with 1 method)

julia> @NLexpression(m, F1, x[1]^2 + x[2]^2)
JuMP.NonlinearExpression(Feasibility problem with:
 * 0 linear constraints
 * 2 variables
Solver is Ipopt,1)

julia> @NLexpression(m, F2, (x[1]-5.0)^2 + (x[2]-5.0)^2)
JuMP.NonlinearExpression(Feasibility problem with:
 * 0 linear constraints
 * 2 variables
Solver is Ipopt,2)

julia> @constraint(m, x .>= bounds_BK1()[1])
2-element Array{JuMP.ConstraintRef{JuMP.Model,JuMP.GenericRangeConstraint{JuMP.GenericAffExpr{Float64,JuMP.Variable}}},1}:
 x[1] ≥ -5
 x[2] ≥ -5

julia> @constraint(m, x .<= bounds_BK1()[2])
2-element Array{JuMP.ConstraintRef{JuMP.Model,JuMP.GenericRangeConstraint{JuMP.GenericAffExpr{Float64,JuMP.Variable}}},1}:
 x[1] ≤ 10
 x[2] ≤ 10

julia> s = solve(m)

******************************************************************************
This program contains Ipopt, a library for large-scale nonlinear optimization.
Ipopt is released as open source code under the Eclipse Public License (EPL).
         For more information visit http://projects.coin-or.org/Ipopt
******************************************************************************

:Optimal

julia> getvalue(x)
2-element Array{Float64,1}:
 0.261454
 0.261454

I do not believe that it is currently possible for you to build up an array of expressions within your functs_BK1 that would later be passed to @NLexpression. At present, one needs to pass the scalar expressions of JuMP variables directly in as arguments to @NLexpression.

JuMP - MethodError: `getindex` has no method matching getindex(::Function, ::Int64)

1 Answers1