Oscar Dowson

A few updates for this: * The BanditDuality is implemented: https://github.com/odow/SDDP.jl/blob/23fcaa9085e6ec43e1e0423b4e64c078951e8101/src/plugins/duality_handlers.jl#L310-L412 * I spent a fair bit of time getting a better understanding of the Lagrangian problem. We now just...

What is `import Pkg; Pkg.status()`? It looks like you've installed an old version somehow.

Okay, you're running some pretty old versions there! Do ```Julia import Pkg Pkg.rm("MathOptFormat") Pkg.update() ``` The latest version of SDDP.jl is v0.4.4. I think the problem is that you have...

Also, please update to Julia 1.6. We're dropping support for Julia 1.5, but the good news is that Julia 1.6 is the new long-term support release of Julia, so you'll...

The error in the example you sent me is because you had ```Julia @variable(sp, 0

@andrewrosemberg does this seem reasonable? Is it easy to build the forward and backward models with the same state variables?

Very nice. Don't let the `jump-dev` repos block the release. We can take care to update them in the future. They aren't urgent.

I'll post the same comment here as from Discourse: The “numerical instability” issue is coming from GLPK. I’ve seen it frequently in cutting plane type problems. Try a different solver:...

Can documenter show gifs? This could be a cool tutorial to include.

```Julia import Ipopt const MOI = Ipopt.MOI for ls in ["ma97", "pardiso", "mumps"] model = Ipopt.Optimizer() MOI.set(model, MOI.RawOptimizerAttribute("linear_solver"), ls) x = MOI.add_variable(model) MOI.set(model, MOI.ObjectiveSense(), MOI.MIN_SENSE) f = 1.0 * x...