Oscar Dowson

This doesn't touch the problematic `JuMP = "=1.2.1"` Okay... this obvious needs a few improvements.

Do not use SSH key to upload documentation

2

This shouldn't be necessary

Add ability to reset basis matrix

2

From a discussion with @joaquimg: Add HiGHS options for resetting the basis and refactoring the matrix, so that the model is in a "just been built" state without actually needing...

Use plot recipe instead of pyplot

9

Pyplot is a pretty heavy dependency. An alternative is to write a plotting recipe: https://github.com/JuliaPlots/RecipesBase.jl This lets you have a very light dependency, and users can use whichever Plots.jl backed...

Choosing lattice branching structure

1

So the lattice approximation is very cool. It's looking promising that I can integrate this into SDDP.jl. Instead of giving a fixed structure for each stage, is there any research...

PublicationPlots do not check for non-finite data

2

From an un-named person on WhatsApp:

Fix use of private symbol in MOI.FileFormats.MPS

1

https://github.com/jump-dev/MathOptInterface.jl/pull/2421 broke BilevelJuMP. This will fix it, but I'll also revert the relevant portion of the change in MOI.

kHighsModelStatusOptimal reported for non-convex QP

1

Solving `Max x * y` returns "Optimal" instead of declaring the QP to be non-convex. See the JuMP example: ```julia julia> using JuMP, HiGHS julia> model = Model(HiGHS.Optimizer) A JuMP...

Add @nonlinear macro for modifying how expressions are parsed

7

x-ref https://github.com/jump-dev/JuMP.jl/issues/3498 x-ref https://github.com/jump-dev/JuMP.jl/issues/3729 - [x] I'm open to bikeshedding the name (and whether we should even add this). A core problem is that @mitchphillipson's models from #3729 contain `(x...

Nonlinear subexpressions

4

Background * JuMP has long supported nonlinear subexpressions created with `@NLexpression`. * The AD system in `MOI.Nonlinear` has support for representing expressions, and dealing with them efficiently * When we...