search

SciML / NonlinearSolve.jl

High-performance and differentiation-enabled nonlinear solvers (Newton methods), bracketed rootfinding (bisection, Falsi), with sparsity and Newton-Krylov support.
https://docs.sciml.ai/NonlinearSolve/stable/
MIT License
221 stars 39 forks source link

NonlinearSolve v3 Release #302

Closed avik-pal closed 7 months ago

avik-pal commented 7 months ago

Wait on https://github.com/SciML/NonlinearSolve.jl/pull/297, https://github.com/SciML/NonlinearSolve.jl/pull/303

Fixes #301, Fixes #300, Fixes #298, Fixes #293, Fixes #255

@ChrisRackauckas we need to update the compat entries for SimpleNonlinearSolve in OrdinaryDiffEq, ModelingToolkit, etc. for the doc build to go through

Main Changes

  1. Support only SimpleNonlinearSolve 1+
  2. Drop the old ad specification; only support a direct autodiff
  3. Update all documentation
  4. Pulled in some of the doc changes from #276
  5. If the problem uses StaticArrays, we try to use a Simple* algorithm first for the non-caching solve

TODOs

codecov[bot] commented 7 months ago

Codecov Report

Attention: 22 lines in your changes are missing coverage. Please review.

Comparison is base (b853842) 86.08% compared to head (50cef59) 85.99%.

Files Patch % Lines
ext/NonlinearSolveMINPACKExt.jl 77.50% 9 Missing :warning:
ext/NonlinearSolveNLsolveExt.jl 86.66% 6 Missing :warning:
src/extension_algs.jl 69.23% 4 Missing :warning:
src/broyden.jl 88.88% 1 Missing :warning:
src/default.jl 87.50% 1 Missing :warning:
src/klement.jl 88.88% 1 Missing :warning:
Additional details and impacted files ```diff @@ Coverage Diff @@ ## master #302 +/- ## ========================================== - Coverage 86.08% 85.99% -0.09% ========================================== Files 21 23 +2 Lines 1825 1935 +110 ========================================== + Hits 1571 1664 +93 - Misses 254 271 +17 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

avik-pal commented 7 months ago

@ChrisRackauckas the docs build is working locally with https://github.com/SciML/OrdinaryDiffEq.jl/pull/2076. The GPU failure which we are seeing on master needs a new tag for Adapt https://github.com/JuliaGPU/Adapt.jl/issues/74. Finally #300, #301, #276 have been pulled in here

maleadt commented 7 months ago

The GPU failure which we are seeing on master needs a new tag for Adapt JuliaGPU/Adapt.jl#74

https://github.com/JuliaRegistries/General/pull/96765