Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • 15:23

    dpo on gh-pages

    build based on 4cc053c (compare)

  • 15:20

    abelsiqueira on jso-docs

    (compare)

  • 15:20

    abelsiqueira on main

    Move tutorials to jso-docs (compare)

  • 15:20
    abelsiqueira closed #81
  • 13:31
    dpo commented #225
  • 13:28
    paraynaud commented #225
  • 13:18
    paraynaud commented #225
  • 12:59
    paraynaud commented #225
  • 12:59
    paraynaud commented #225
  • 12:24
    dpo commented #225
  • Jun 24 20:19
    paraynaud edited #235
  • Jun 24 20:19
    paraynaud edited #235
  • Jun 24 20:19
    paraynaud edited #235
  • Jun 24 20:18
    paraynaud edited #235
  • Jun 24 20:18
    paraynaud commented #235
  • Jun 24 20:12
    paraynaud commented #235
  • Jun 24 20:11
    paraynaud opened #235
  • Jun 24 18:59
    paraynaud commented #225
  • Jun 24 17:16

    dpo on gh-pages

    build based on 738a309 (compare)

  • Jun 24 17:14
    dpo commented #599
Christopher Rackauckas
@ChrisRackauckas
hey!
Abel Soares Siqueira
@abelsiqueira
Hello
Christopher Rackauckas
@ChrisRackauckas
I'm just poking around
interested in how it's different from JuMP
all of my models are nonlinear and so it was never truly a pleasure to work with in that domain
Abel Soares Siqueira
@abelsiqueira
I replied on discourse, but it is in moderation
It's basically what Miles said
This is focused in Nonlinear Programming
but also in writing solvers
Depending on how you use your models, it might be worth investigating
Christopher Rackauckas
@ChrisRackauckas
Usually doing it for parameter estimation of differential equations
so something where the objective function does not have an analytical solution itself
and it may be unconstrained optimization
I am looking to target the simplest interface to hit as many Julia packages as possible
that's what I liked about JuMP, but it's not very nonlinear-focused
Abel Soares Siqueira
@abelsiqueira
JuMP really is great for creating models
We do not compete in that category
What we give is the possibility for someone wanting more control, like entering the gradient and Hessian by hand
Or for people wanting to use CUTEst
Christopher Rackauckas
@ChrisRackauckas
oh I see
fadihamad94
@fadihamad94
Hello, I am trying to do some benchmarking. I want to run CUTEst test set with Ipopt optimizer. Can anyone help if possible.
The issue is that Ipopt.jl can now be execute either through JuMP or through MathOptInterface
we use to do that using MathProgBase, but this is not supported anymore
Dominique
@dpo
Hi @fadihamad94. You can use NLPModelsIpopt.jl, which is a thin interface to Ipopt.jl with support for the NLPModels API. Among others, you'll be able to pass CUTEst.jl problems. Please let us know if you have any questions.