Test master #3094

ChrisRackauckas · 2024-10-05T14:14:20Z

No description provided.

github-actions · 2024-10-05T14:15:30Z

README.md

@@ -24,6 +24,7 @@ For information on using the package,
 [in-development documentation](https://docs.sciml.ai/ModelingToolkit/dev/) for the version of
 the documentation which contains the unreleased features.

+


[JuliaFormatter] _{reported by reviewdog 🐶}

Suggested change

ChrisRackauckas · 2024-10-05T15:02:01Z

@AayushSabharwal is this what you were talking about the other day? https://github.com/SciML/ModelingToolkit.jl/actions/runs/11193882959/job/31119719069?pr=3094#step:6:1125

hersle · 2024-10-05T15:54:54Z

It broke from the 3rd to the 1st commit between Sept. 27-28, where only DiffEqCallbacks was bumped to v4. Reverting to DiffEqCallbacks v3 does not fix it on my machine, so I think the cause is external. ~~Maybe something solver-related in this test call?~~

AayushSabharwal · 2024-10-05T16:00:07Z

@AayushSabharwal is this what you were talking about the other day?

Yes

Maybe something solver-related in this test call?

That test call fails because it uses the tstops from the previous solve, which also fails.

hersle · 2024-10-05T16:42:07Z

SciML/DiffEqBase.jl@f870554 breaks the test.

ChrisRackauckas · 2024-10-05T16:53:49Z

Really? That's related????

ChrisRackauckas · 2024-10-05T16:57:29Z

Just decrease the tolerance, that's effectively random.

hersle · 2024-10-05T17:00:51Z

Yeah, it seems very sensitive to the tolerance. I just did checkout on some commits until this hit.

hersle · 2024-10-05T17:31:14Z

using Printf
for pow in range(-10, -3)
    for fac in range(1, 9)
        reltol = fac * 10.0^pow
        sol1 = solve(prob, RosShamp4(); reltol)
        @printf "reltol = %.1E -> %s\n" reltol sol1.retcode
    end
end

after f870554                   before f870554
============================    ============================
reltol = 1.0E-10 -> Unstable    reltol = 1.0E-10 -> Unstable
reltol = 2.0E-10 -> Unstable    reltol = 2.0E-10 -> Unstable
reltol = 3.0E-10 -> Success     reltol = 3.0E-10 -> Success
reltol = 4.0E-10 -> Unstable    reltol = 4.0E-10 -> Unstable
reltol = 5.0E-10 -> Unstable    reltol = 5.0E-10 -> Unstable
reltol = 6.0E-10 -> Unstable    reltol = 6.0E-10 -> Unstable
reltol = 7.0E-10 -> Unstable    reltol = 7.0E-10 -> Unstable
reltol = 8.0E-10 -> Unstable    reltol = 8.0E-10 -> Unstable
reltol = 9.0E-10 -> Unstable    reltol = 9.0E-10 -> Unstable
reltol = 1.0E-09 -> Unstable    reltol = 1.0E-09 -> Unstable
reltol = 2.0E-09 -> Unstable    reltol = 2.0E-09 -> Unstable
reltol = 3.0E-09 -> Unstable    reltol = 3.0E-09 -> Unstable
reltol = 4.0E-09 -> Unstable    reltol = 4.0E-09 -> Unstable
reltol = 5.0E-09 -> Unstable    reltol = 5.0E-09 -> Unstable
reltol = 6.0E-09 -> Unstable    reltol = 6.0E-09 -> Unstable
reltol = 7.0E-09 -> Unstable    reltol = 7.0E-09 -> Unstable
reltol = 8.0E-09 -> Unstable    reltol = 8.0E-09 -> Unstable
reltol = 9.0E-09 -> Unstable    reltol = 9.0E-09 -> Unstable
reltol = 1.0E-08 -> Unstable    reltol = 1.0E-08 -> Unstable
reltol = 2.0E-08 -> Unstable    reltol = 2.0E-08 -> Unstable
reltol = 3.0E-08 -> Unstable    reltol = 3.0E-08 -> Unstable
reltol = 4.0E-08 -> Unstable    reltol = 4.0E-08 -> Unstable
reltol = 5.0E-08 -> Unstable    reltol = 5.0E-08 -> Unstable
reltol = 6.0E-08 -> Unstable    reltol = 6.0E-08 -> Unstable
reltol = 7.0E-08 -> Unstable    reltol = 7.0E-08 -> Unstable
reltol = 8.0E-08 -> Unstable    reltol = 8.0E-08 -> Unstable
reltol = 9.0E-08 -> Unstable    reltol = 9.0E-08 -> Unstable
reltol = 1.0E-07 -> Unstable    reltol = 1.0E-07 -> Unstable
reltol = 2.0E-07 -> Unstable    reltol = 2.0E-07 -> Unstable
reltol = 3.0E-07 -> Unstable    reltol = 3.0E-07 -> Unstable
reltol = 4.0E-07 -> Unstable    reltol = 4.0E-07 -> Unstable
reltol = 5.0E-07 -> Unstable    reltol = 5.0E-07 -> Unstable
reltol = 6.0E-07 -> Unstable    reltol = 6.0E-07 -> Unstable
reltol = 7.0E-07 -> Unstable    reltol = 7.0E-07 -> Unstable
reltol = 8.0E-07 -> Success     reltol = 8.0E-07 -> Unstable
reltol = 9.0E-07 -> Unstable    reltol = 9.0E-07 -> Unstable
reltol = 1.0E-06 -> Unstable    reltol = 1.0E-06 -> Unstable
reltol = 2.0E-06 -> Unstable    reltol = 2.0E-06 -> Unstable
reltol = 3.0E-06 -> Unstable    reltol = 3.0E-06 -> Unstable
reltol = 4.0E-06 -> Unstable    reltol = 4.0E-06 -> Unstable
reltol = 5.0E-06 -> Success     reltol = 5.0E-06 -> Unstable
reltol = 6.0E-06 -> Unstable    reltol = 6.0E-06 -> Unstable
reltol = 7.0E-06 -> Success     reltol = 7.0E-06 -> Success
reltol = 8.0E-06 -> Unstable    reltol = 8.0E-06 -> Unstable
reltol = 9.0E-06 -> Unstable    reltol = 9.0E-06 -> Success
reltol = 1.0E-05 -> Unstable    reltol = 1.0E-05 -> Unstable
reltol = 2.0E-05 -> Success     reltol = 2.0E-05 -> Success
reltol = 3.0E-05 -> Success     reltol = 3.0E-05 -> Success
reltol = 4.0E-05 -> Unstable    reltol = 4.0E-05 -> Unstable
reltol = 5.0E-05 -> Unstable    reltol = 5.0E-05 -> Unstable
reltol = 6.0E-05 -> Success     reltol = 6.0E-05 -> Success
reltol = 7.0E-05 -> Unstable    reltol = 7.0E-05 -> Unstable
reltol = 8.0E-05 -> Unstable    reltol = 8.0E-05 -> Unstable
reltol = 9.0E-05 -> Success     reltol = 9.0E-05 -> Unstable
reltol = 1.0E-04 -> Success     reltol = 1.0E-04 -> Success
reltol = 2.0E-04 -> Success     reltol = 2.0E-04 -> Success
reltol = 3.0E-04 -> Success     reltol = 3.0E-04 -> Success
reltol = 4.0E-04 -> Success     reltol = 4.0E-04 -> Unstable
reltol = 5.0E-04 -> Success     reltol = 5.0E-04 -> Success
reltol = 6.0E-04 -> Unstable    reltol = 6.0E-04 -> Unstable
reltol = 7.0E-04 -> Success     reltol = 7.0E-04 -> Success
reltol = 8.0E-04 -> Unstable    reltol = 8.0E-04 -> Unstable
reltol = 9.0E-04 -> Success     reltol = 9.0E-04 -> Success
reltol = 1.0E-03 -> Success     reltol = 1.0E-03 -> Success
reltol = 2.0E-03 -> Success     reltol = 2.0E-03 -> Success
reltol = 3.0E-03 -> Success     reltol = 3.0E-03 -> Success
reltol = 4.0E-03 -> Success     reltol = 4.0E-03 -> Success
reltol = 5.0E-03 -> Success     reltol = 5.0E-03 -> Success
reltol = 6.0E-03 -> Success     reltol = 6.0E-03 -> Success
reltol = 7.0E-03 -> Success     reltol = 7.0E-03 -> Success
reltol = 8.0E-03 -> Success     reltol = 8.0E-03 -> Success
reltol = 9.0E-03 -> Unstable    reltol = 9.0E-03 -> Unstable

AayushSabharwal · 2024-10-08T06:44:35Z

I picked a tolerance from @hersle 's table above at the beginning of what looks like a relatively stable region

AayushSabharwal · 2024-10-08T09:53:53Z

Huh, that passed locally

AayushSabharwal · 2024-10-08T10:17:22Z

So far no luck in replicating any of the failures in InterfaceI

hersle · 2024-10-08T10:21:06Z

I got the feeling the test is not robust regardless, considering how SciML/DiffEqBase.jl@f870554 breaks it. Maybe even machine-dependent floating point operations etc. play a role.

AayushSabharwal · 2024-10-08T10:26:59Z

That's possible, but how is it getting differently sized arrays given the tstops of the previous solve?

AayushSabharwal · 2024-10-08T10:27:39Z

Hmm, there's this warning:

┌ Warning: At t=0.6261972386805764, dt was forced below floating point epsilon 1.1102230246251565e-16, and step error estimate = 3.138704824490307. Aborting. There is either an error in your model specification or the true solution is unstable (or the true solution can not be represented in the precision of Float64).
└ @ SciMLBase ~/.julia/packages/SciMLBase/PTTHz/src/integrator_interface.jl:609

right before the failure. Maybe the first solve is failing on CI?

hersle · 2024-10-08T10:34:21Z

I think that's what causes it.

AayushSabharwal · 2024-10-08T11:52:11Z

I also fixed a failure in InterfaceII's dde tests, but it's failing differently now and is also not reproducing locally. This is really weird behavior

EDIT: Okay never mind I guess

Test master

e926d06

github-actions bot reviewed Oct 5, 2024

View reviewed changes

Update tearing.jl

eb8b617

ChrisRackauckas and others added 3 commits October 5, 2024 15:21

Update tearing.jl

b57f7f3

fix: fix test

ce64618

refactor: format

f632576

AayushSabharwal added 2 commits October 8, 2024 13:03

test: fix depwarn in dde test

b160316

test: decrease solve tolerance, increase test tolerance

7aac0fd

AayushSabharwal mentioned this pull request Oct 8, 2024

fix: detect observed variables and dependent parameters dependent on discrete parameters #3106

Open

5 tasks

AayushSabharwal added 2 commits October 9, 2024 11:24

test: add REPL as a test dependency

03c3861

test: remove ODAEProblem test

422230b

ChrisRackauckas merged commit 725da10 into master Oct 9, 2024
21 of 26 checks passed

ChrisRackauckas deleted the ChrisRackauckas-patch-6 branch October 9, 2024 12:09

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Test master #3094

Test master #3094

ChrisRackauckas commented Oct 5, 2024

github-actions bot Oct 5, 2024

ChrisRackauckas commented Oct 5, 2024

hersle commented Oct 5, 2024 •

edited

Loading

AayushSabharwal commented Oct 5, 2024 •

edited

Loading

hersle commented Oct 5, 2024

ChrisRackauckas commented Oct 5, 2024

ChrisRackauckas commented Oct 5, 2024

hersle commented Oct 5, 2024

hersle commented Oct 5, 2024 •

edited

Loading

AayushSabharwal commented Oct 8, 2024

AayushSabharwal commented Oct 8, 2024

AayushSabharwal commented Oct 8, 2024

hersle commented Oct 8, 2024

AayushSabharwal commented Oct 8, 2024

AayushSabharwal commented Oct 8, 2024

hersle commented Oct 8, 2024

AayushSabharwal commented Oct 8, 2024 •

edited

Loading

		@@ -24,6 +24,7 @@ For information on using the package,
		[in-development documentation](https://docs.sciml.ai/ModelingToolkit/dev/) for the version of
		the documentation which contains the unreleased features.

Test master #3094

Test master #3094

Conversation

ChrisRackauckas commented Oct 5, 2024

github-actions bot Oct 5, 2024

Choose a reason for hiding this comment

ChrisRackauckas commented Oct 5, 2024

hersle commented Oct 5, 2024 • edited Loading

AayushSabharwal commented Oct 5, 2024 • edited Loading

hersle commented Oct 5, 2024

ChrisRackauckas commented Oct 5, 2024

ChrisRackauckas commented Oct 5, 2024

hersle commented Oct 5, 2024

hersle commented Oct 5, 2024 • edited Loading

AayushSabharwal commented Oct 8, 2024

AayushSabharwal commented Oct 8, 2024

AayushSabharwal commented Oct 8, 2024

hersle commented Oct 8, 2024

AayushSabharwal commented Oct 8, 2024

AayushSabharwal commented Oct 8, 2024

hersle commented Oct 8, 2024

AayushSabharwal commented Oct 8, 2024 • edited Loading

hersle commented Oct 5, 2024 •

edited

Loading

AayushSabharwal commented Oct 5, 2024 •

edited

Loading

hersle commented Oct 5, 2024 •

edited

Loading

AayushSabharwal commented Oct 8, 2024 •

edited

Loading