Jean-Baptiste Caillau

@SobhanMP @alexQueue hi there, do I have to understand that using `Int64` indices solves the problem?

Hi, same `no method matching operation(::Symbolics.TermCombination)` error in [OptimalControl.jl](https://github.com/control-toolbox/OptimalControl.jl) and related packages that rely on [ADNLModels.jl](https://github.com/JuliaSmoothOptimizers/ADNLPModels.jl), the latter using Symbolics.jl for sparsity pattern analysis (@tmigot and @amontoison can be more...

Hi @tmigot and @amontoison (again) Thanks for the meeting. Below a few things I noted. The current bottleneck seems to be the sparse Hessian (of the Lagrangian in the case...

@amontoison @tmigot amazing: what would you guys say is the game changer here?

@sshin23 @frapac hi there, on a related matter: when declaring (lower and upper) bounds on ExaModels variables, is it equivalent (= are the generated models eventually identical) to - declare...

> [@jbcaillau](https://github.com/jbcaillau) how to handle constraint (via variable bound or separate constraint) is up to the solver. Ipopt and MadNLP can be more efficient when you declare them as variable...

@sshin23 sure. following use case: imagine the user declares ```julia x = variable(c) # then later constraint(c, x; lcon = -1, ucon = 1) ``` i assume it is always...

> [@jbcaillau](https://github.com/jbcaillau) I see. I'd imagine you want to reset some bound values after solving the problem? for instance that, yes > Would something like this work? > > variable!(x,...

@sshin23 more precisely, imagine that the user declares ```julia x = ExaModels.variable(c, n, 1:N+1; start=xs) # n >= 2 # then later ExaModels.constraint(c, x[2, j] for j ∈ 1:N+1; lcon=0,...

thanks @sshin23 ; not a matter a preference but of sequence of actions. indeed, would be nice to be able to *update* bounds associated with a variable, *e.g.* with your...