Write Solution to a .SOL File

[stumarcus314]

Is it possible to have SCIP.jl write the solution to a .SOL file by setting an optimizer attribute?

m = JuMP.Model()
[...]
    JuMP.set_optimizer(m, SCIP.Optimizer)
    JuMP.set_optimizer_attribute(m, "SolutionFile", "result.sol")
    JuMP.optimize!(m)

[matbesancon]

Not directly, but you can use SCIPprintBestSol on the SCIP.Optimizer See https://stackoverflow.com/questions/43399199/how-can-i-store-the-results-into-a-text-file for the SCIP function API. As an alternative, you can extract the values from JuMP and produce the SOL file yourself, for example looping over all variables and adding the name and value to your file.

…

On Tue, Feb 25, 2025, 18:54 stumarcus314 ***@***.***> wrote: Is it possible to have SCIP.jl write the solution to a .SOL file by setting an optimizer attribute? m = JuMP.Model() [...] JuMP.set_optimizer(m, SCIP.Optimizer) JuMP.set_optimizer_attribute(m, "SolutionFile", "result.sol") JuMP.optimize!(m) — Reply to this email directly, view it on GitHub <#302>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB2FDMQ4VZMM6JDTURZNVJT2RSU57AVCNFSM6AAAAABX3IWSBGVHI2DSMVQWIX3LMV43ASLTON2WKOZSHA3TSMJWGU2TGMA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***> [image: stumarcus314]*stumarcus314* created an issue (scipopt/SCIP.jl#302) <#302> Is it possible to have SCIP.jl write the solution to a .SOL file by setting an optimizer attribute? m = JuMP.Model() [...] JuMP.set_optimizer(m, SCIP.Optimizer) JuMP.set_optimizer_attribute(m, "SolutionFile", "result.sol") JuMP.optimize!(m) — Reply to this email directly, view it on GitHub <#302>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB2FDMQ4VZMM6JDTURZNVJT2RSU57AVCNFSM6AAAAABX3IWSBGVHI2DSMVQWIX3LMV43ASLTON2WKOZSHA3TSMJWGU2TGMA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[stumarcus314]

Can you provide an example using JuMP how to call SCIPprintBestSol on the SCIP.Optimizer?

[matbesancon]

m = Model(SCIP.Optimizer)
# build problem
optimize!(m)
open("mysol.sol", "w") do f
           SCIP.LibSCIP.SCIPprintBestSol(JuMP.backend(m).optimizer.model, Libc.FILE(f), true)
end

[odow]

Closing as won't fix because this is not something I want to provide access to via MOI.

The example can be simplified with the use of JuMP.unsafe_backend(model):

julia> using JuMP, SCIP

julia> model = Model(SCIP.Optimizer)
A JuMP Model
├ solver: SCIP
├ objective_sense: FEASIBILITY_SENSE
├ num_variables: 0
├ num_constraints: 0
└ Names registered in the model: none

julia> @variable(model, x >= 0)
x

julia> optimize!(model)
feasible solution found by trivial heuristic after 0.0 seconds, objective value 0.000000e+00
presolving:
presolving (1 rounds: 1 fast, 0 medium, 0 exhaustive):
 1 deleted vars, 0 deleted constraints, 0 added constraints, 0 tightened bounds, 0 added holes, 0 changed sides, 0 changed coefficients
 0 implications, 0 cliques
transformed 1/3 original solutions to the transformed problem space
Presolving Time: 0.00

SCIP Status        : problem is solved [optimal solution found]
Solving Time (sec) : 0.00
Solving Nodes      : 0
Primal Bound       : +0.00000000000000e+00 (3 solutions)
Dual Bound         : +0.00000000000000e+00
Gap                : 0.00 %

julia> open("output.sol", "w") do io
           SCIP.LibSCIP.SCIPprintBestSol(unsafe_backend(model), Libc.FILE(io), true)
           return
       end

shell> cat output.sol
objective value:                                    0
x                                               50000 	(obj:0)