demand function is not working

[floswald]

this call is outdated

https://github.com/ScPo-CompEcon/CourseMatch.jl/blob/master/src/demand.jl#L117

update to new API. you need to do ] up in teh coursematch project

[jeannesorin]

Does m = Model(with_optimizer(Ipopt.Optimizer))work or is there an alternative for CbcSolver? I could not find anything else online.

[floswald]

   _       _ _(_)_     |  Documentation: https://docs.julialang.org
  (_)     | (_) (_)    |
   _ _   _| |_  __ _   |  Type "?" for help, "]?" for Pkg help.
  | | | | | | |/ _` |  |
  | | |_| | | | (_| |  |  Version 1.1.0 (2019-01-21)
 _/ |\__'_|_|_|\__'_|  |  Official https://julialang.org/ release
|__/                   |

julia> using JuMP

julia> using Cbc

julia> m = Model(with_optimizer(Cbc.Optimizer))
A JuMP Model
Feasibility problem with:
Variables: 0
Model mode: AUTOMATIC
CachingOptimizer state: EMPTY_OPTIMIZER
Solver name: COIN Branch-and-Cut (Cbc)

[floswald]

anyone working on this?

[danielbgyetvai]

Yes, I am. I have updated the constraint calls in demand!() and am now trying to figure out how to adapt the objective.

[floswald]

excellent! keep us posted with how it goes / any questions.

[danielbgyetvai]

It is not directly relevant but I created the following function (replacing and adapting rand_constraints() which was not working) in student.jl:

#This is an auxiliary function for testing purposes only. It might create inconsistencies, economically speaking.
function add_rand_constraints!(s::Student)
    N = length(s.constraints.isFC)
    if N<2
        error("N must be > 2")
    end
    s.constraints.mandatory = rand([true false],N)
    s.constraints.notTCprogram = rand([true false false false false false],N)
    s.constraints.notTCsemester = rand([true false false false false false],N)
    s.constraints.isTC = rand([true false],N)
    s.constraints.notFCprogram = rand([true false false false false false],N)
    s.constraints.notFCsemester = rand([true false false false false false],N)
    s.constraints.isFC = rand([true false],N)
    s.constraints.notELprogram = rand([true false false false false false],N)
    s.constraints.notELsemester = rand([true false false false false false],N)
    s.constraints.isEL = rand([true false],N)
end 

This can still be improved since I feel it might create economically inconsistent sets of constraints like courses for which isTC is 0 and notTCprogram is 0 as well.

[gregorydannay]

Hey floswald,

We have some troubles with the solver. We manage to change all the constraints but the Cbc solver does not handle NL objective function. We tried with Ipopt but it doesn't handle mixed integers. Do you know a solver that can solve Mixed Integers Non Linear Programming problems ?