Help with Performance in Large Mixed Complementarity Problem

Don’t worry, I’ve figured it out. It means that PATH has errored. The bug is that we don’t catch this properly in the JuMP interface.

It means that one of the following is true:

You have a license issue:

there are no variables

or there are too many non-zeros

Hard to say more without a reproducible example.

1 Like

Sorry, I forgot the first line

ERROR: LoadError: MethodError: Cannot `convert` an object of type Nothing to an object of type Vector{Float64}

1 Like

Thanks @odow my bet is that it is a license issue because the code runs in other machine.

However, what I find puzzling is that the simple example on PATHsolvers.jl does run.

1 Like

However, what I find puzzling is that the simple example

You can solve problems with up to 300 variables without a license.

See GitHub - chkwon/PATHSolver.jl: provides a Julia wrapper for the PATH Solver for solving mixed complementarity problems for details on how to set the license.

I’ll fix the bug today: Fix MOI wrapper if solve_mcp fails by odow · Pull Request #101 · chkwon/PATHSolver.jl · GitHub