Oscar Dowson

> I can go ahead and alter my PR to implement that Sure. The lines to change are: https://github.com/JuliaDocs/Documenter.jl/blob/056284a9850a0e74a439a694e267aff20a7311f5/src/DocTests.jl#L198-L206

So far it looks like your time limit is 100 seconds, but the time has only run for 8.8 seconds?

Hmm. We did: https://github.com/JuliaRegistries/General/pull/57306 https://github.com/JuliaRegistries/General/blob/407a928e4790630774897f4facfd14a0580584d2/J/Juniper/Versions.toml#L70-L71 But it's not showing up as a release https://github.com/lanl-ansi/Juniper.jl/releases

From my point of view (I maintain the current binaries https://github.com/ERGO-Code/HiGHS#download as part of the JuMP project in Julia and am not part of the HiGHS team): Given the limited...

```Julia using SDDP import Ipopt function main() δ, ρ = 0.1, 0.05 β = 1 / (1 + ρ) graph = SDDP.LinearGraph(1) SDDP.add_edge(graph, 1 => 1, β) model = SDDP.PolicyGraph(...

We don't support nonlinear objectives just yet. You'll need to add an epigraph variable. ```Julia model = SDDP.PolicyGraph( graph; sense = :Max, upper_bound = 1000, optimizer = opt, ) do...

Oh. `@NLconstraint(model, t

I should have thought about this a fraction more. The issue now is that `log(0)` is undefined, so it breaks when it explores the `u = 0` action. Add a...

> it's kinda slow (requires many iterations for accurate sol...). I bet my code could be optimized a lot... And/Or maybe I need to use a better solver. Three parts...

But just for a ballpark figure, if there are 1000 state variables, and each state has two potential values (on/off), then there are 2^1000 (~10^300) values in the state-space, and...