Skip to content

Latest commit

 

History

History
31 lines (28 loc) · 2.14 KB

roadmap.md

File metadata and controls

31 lines (28 loc) · 2.14 KB

Development roadmap

The JuMP developers have compiled this roadmap document to share their plans and goals with the JuMP community. Contributions to roadmap issues are especially invited.

Most of these issues will require changes to both JuMP and MathOptInterface, and are non-trivial in their implementation. They are in no particular order, but represent broad themes that we see as areas in which JuMP could be improved.

  • Make nonlinear programming a first-class citizen. There have been many issues and discussions about this: currently nonlinear constraints are handled through a MOI.NLPBlock and have various limitations and restrictions.
  • Add support for coefficient types other than Float64: #2025 Since the very beginning, JuMP has hard-coded the coefficient type as Float64. This has made it impossible to support solvers which can use other types such as BigFloat or Rational{BigInt}.
  • Add support for constraint programming: #2227 JuMP has a strong focus on linear, conic and nonlinear optimization problems. We want to add better support for constraint programming.
  • Refactor the internal code of JuMP's macros. The code in src/macros.jl is some of the oldest part of JuMP and is difficult to read, modify, and extend. We should overhaul the internals of JuMP's macros---without making user-visible breaking changes---to improve their long-term maintainability.