You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Releases v3.25.0, v3.25.1 show regressions for OptimizationOptimJL.Optim.BFGS() when used with upper/lower bounds (v3.25.0, see #747) and when used with user-supplied derivative in combination with upper/lower bounds (v3.25.1 #754)
Expected behavior
Ideally tests should exist that catch these problems before release...
Edit - apologies, looks like the real problem here is perhaps how the tests are organized, which makes it difficult to
maintain and add tests in response to issues and fixes?
See #758 (comment)
SciML#754SciML#747
partially addressing SciML#755
NB: unrelated problem, tests using Optimization.AutoModelingToolkit() were failing,
added code
@test (sol = solve(prob, Optim.Newton())) isa Any # test exception not thrown
and similar so exception is caught and other tests run
See
SciML#760SciML#755
- add tests for Optim.BFGS() with user-supplied grad and bounds
- fix wording for error message if bounds specified without either grad function or AD
Describe the bug 🐞
Releases v3.25.0, v3.25.1 show regressions for OptimizationOptimJL.Optim.BFGS() when used with upper/lower bounds (v3.25.0, see #747) and when used with user-supplied derivative in combination with upper/lower bounds (v3.25.1 #754)
Expected behavior
Ideally tests should exist that catch these problems before release...
Edit - apologies, looks like the real problem here is perhaps how the tests are organized, which makes it difficult to
maintain and add tests in response to issues and fixes?
See #758 (comment)
Minimal Reproducible Example 👇
See #747 (comment), #754
Additional context
The text was updated successfully, but these errors were encountered: