Skip to content

Latest commit

 

History

History
162 lines (122 loc) · 5.53 KB

NEWS.md

File metadata and controls

162 lines (122 loc) · 5.53 KB
Raw

Release notes

v0.8.2 (May 15th, 2024)

Bug Fixes

  • [MOI] fix eval_constraint_jacobian_product (#337)
  • [MOI] Add support for nonlinear problems without Hessian (#322)
  • [MadNLPGPU] Bug fix for empty Hessian (#326)

New feature

  • [LinearSolvers] Add support for LDL factorization in CHOLMOD (#321)
  • Add an ordering for cuDSS (#317)

Performance and maintenance

  • Reexport MadNLP from MadNLP/libs (#325)
  • Use GH Action Julia cache (#339)
  • [MAdNLPMumps] sarrays removed (#328)

v0.8.1 (Apr 9th, 2024)

Bug Fixes

  • fix optional arguments in get_index_constraints (#316)
  • CuDSS synchronize added (#314)

Performance and maintenance

  • Introduce linear solvers with version info (#315)
  • Add logos to documentation (#310)

v0.8.0 (Mar 7th, 2024)

Bug Fixes

  • fix non-deterministic behavior by forcing instantiations (#284)

New features

  • LDL factorization improvement (#300)
  • add support for MOI.ScalarNonLinearFunction (#280)
  • [Linear Solver] Added undocumented cholesky solver (#292)
  • ss/init_improve Add supports to CUDSS.jl (#296)
  • ss/cudss_option_cholmod [MOI] add support for MOI.Interval{Float64} (#295)

Performance and maintenance

  • Update README, OPTIONS, CITATION, and documentation (#304)
  • MOI interface moved to ext (#268)
  • Improve kkt creation on GPUs (#299)
  • CompatHelper: add new compat entry for Metis at version 1 for package MadNLPGPU, (keep existing compat) (#298)
  • CompatHelper: add new compat entry for CUDSS at version 0.1 for package MadNLPGPU, (keep existing compat) (#297)
  • update documentation (#293)
  • [Algorithm] Improve LBFGS performance (#290)
  • [API] Expose the options for iterative refinements and quasi-Newton (#287)
  • [README] Fix Options.md typo (#291)
  • [Options] Deactivate scaling if nlp_scaling=false (#289)
  • [API] Simplify arguments of create_kkt_system (#286)
  • [Tests] Remove test nlp_009_010 from MINLPTests tests (#288)
  • Simplify API of SparseCallback and DenseCallback (#285)
  • [MadNLPHSL] Use HSL.jl (#277)
  • add support for CUDA.jl v5 (#283)
  • simplify implementation of kernels (#281)
  • add proper testing for KKT systems (#278)

v0.7.0 (June 12th, 2023)

Performance and maintenance

  • CompatHelper: bump compat for NLPModels to 0.20, (keep existing compat) (#252)
  • CompatHelper: bump compat for NLPModels to 0.20 for package MadNLPTests, (keep existing compat) (#253)
  • [MadNLPGPU] Migrate to KernelAbstractions 0.9 (#258)
  • [MadNLPTests] Remove ADNLPModels in deps (#259)
  • Update README.md for JuMP documentation (#260)
  • Update for julia v1.9 (#261)

v0.6.0 (March 23th, 2023)

New features

  • add dense BFGS and compact LBFGS algorithms (#221)

Performance and maintenance

  • Bump CUDA.jl to 0.4 and drop support for CUDA.jl 0.3 in MadNLPGPU (#248)
  • Add support for ADNLPModels 0.5 in MadNLPTests (#240)

v0.5.2 (February 27th, 2023)

Bug fixes

  • Fix URL in README (#236)
  • Fix invalid return code in MOI.TerminationStatus (#241)
  • Fix potential infinite loop in feasibility restoration (#244)
  • Fix initial multipliers when initial KKT has no solution (#243)

Performance and maintenance

  • Replace broadcast operators by explicit for loops in IPM kernels to reduce time-to-first-solve (#231)
  • Remove allocations in nonlinear callbacks (#230)

v0.5.1 (October 20th, 2022)

New features

  • Update MOI wrapper to match new Ipopt's MOI wrapper (#224, #233)
  • Add new constructor for custom KKT type (#232)

Bug fixes

  • Fix detection of number of upper and lower bounds for optimization variables (#211)
  • Fix solve! function when dual is provided as input (#215)
  • Update initialization of meta field (#216)
  • Fix type stability in MadNLP (#220, #227, #228)

Performance and maintenance

  • Doc: Fix typos in quickstart.md (#210)
  • Test properly solve! function (#219)
  • Improve error messages when invalid number is detected (#226)

v0.5.0 (August 30th, 2022)

Breaking changes

  • Change names to follow JSO conventions
    • AbstractInteriorPointSolver -> AbstractMadNLPSolver
    • InteriorPointSolver -> MadNLPSolver
    • IPMOptions -> MadNLPOptions
    • Counters -> MadNLPCounters
    • Logger -> MadNLPLogger
  • Linear solvers are now defined as a struct, not a module
    • MadNLPLapackCPU.Solver -> LapackCPUSolver
    • MadNLPUmfpack.Solver -> UmfpackSolver
    • MadNLPLapackGPU.Solver -> LapackGPUSolver
    • MadNLPMa27.Solver -> Ma27Solver
    • MadNLPMa57.Solver -> Ma57Solver
    • MadNLPMa77.Solver -> Ma77Solver
    • MadNLPMa86.Solver -> Ma86Solver
    • MadNLPMa97.Solver -> Ma97Solver
    • MadNLPKrylov.Solver -> KrylovIterator
    • MadNLPMumps.Solver -> MumpsSolver
    • MadNLPPardiso.Solver -> PardisoSolver
    • MadNLPPardisoMKL.Solver -> PardisoMKLSolver
  • Refactor the way we pass options to MadNLP
    • Before: ips = MadNLP.InteriorPointSolver(nlp; option_dict=sparse_options)
    • Now: ips = MadNLP.MadNLPSolver(nlp; sparse_options...)

New features

  • Add support for Float32
  • Add support for NLSModels
  • Add a function timing_madnlp to decompose the time spent in the callbacks and in the linear solver at each iteration
  • Add a CuMadNLPSolver constructor to instantiate MadNLP on CUDA GPU

Bug fixes

  • Stop overwriting the number of threads used in BLAS
  • Use symmetric mul! when using DENSE_CONDENSED_KKT_SYSTEM on the GPU
  • Fix unscaling of the constraints during post-processing

Performance and maintenance

  • Add a toy nonlinear problem HS15Model in MadNLPTests
  • Improve the build of MadNLPHSL
  • Remove StrideOneVector alias in MadNLP
  • Update the documentation
  • Improve code coverage