Fix JAX test failures related to diffrax/optimistix release

[BSnelling]

No description provided.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 78.22%. Comparing base (1c5d990) to head (11b9c90).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #3129       +/-   ##
===========================================
+ Coverage   45.38%   78.22%   +32.83%     
===========================================
  Files         309      315        +6     
  Lines       20216    20482      +266     
  Branches     1484     1484               
===========================================
+ Hits         9176    16023     +6847     
+ Misses      11008     4451     -6557     
+ Partials       32        8       -24     

Flag	Coverage Δ
cpp	71.69% <ø> (?)
cpp_python	37.85% <ø> (ø)
petab	46.56% <ø> (+10.84%)	⬆️
petab_sciml	13.93% <ø> (?)
python	70.35% <ø> (+31.85%)	⬆️
sbmlsuite-jax	33.41% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
python/sdist/amici/jax/petab.py	94.00% <ø> (+75.47%)	⬆️

... and 233 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[socket-security]

Review the following changes in direct dependencies. Learn more about Socket for GitHub.

Diff	Package	Supply Chain Security	Vulnerability	Quality	Maintenance	License
	optax@​0.2.7 ⏵ 0.2.6

View full report

[BSnelling]

Opened #3130 to keep track of this.

[dweindl]

Is it clear why now so many more steps are required? Either way, thanks for fixing that.

[BSnelling]

Is it clear why now so many more steps are required?

Not completely clear. The diffrax update included changes to the linear interpolation timestepper, which could have fairly broad effects: patrick-kidger/diffrax@v0.7.1...v0.7.2

[FFroehlich]

Is it clear why now so many more steps are required?

Not completely clear. The diffrax update included changes to the linear interpolation timestepper, which could have fairly broad effects: patrick-kidger/diffrax@v0.7.1...v0.7.2

I suspect this has something to do with the optimistix update, which is much bigger in scope (patrick-kidger/optimistix@v0.0.11...v0.1.0)

I the future it would be great to track performance across jax/diffrax/optimistix versions.