Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Gradient tests for benchmark collection #967

Closed
yannikschaelte opened this issue Feb 20, 2020 · 4 comments
Closed

Gradient tests for benchmark collection #967

yannikschaelte opened this issue Feb 20, 2020 · 4 comments
Labels
ci enhancement

Comments

@yannikschaelte
Copy link
Member

A lot can go wrong with wrong parameter scales, thus there should be gradient tests for (some of) the benchmark models. Suggest FDs, maybe at (deterministically sampled) random parameters (as the gradients at the optimum tend to be rather sensitive).
@yannikschaelte yannikschaelte added the new Newly created label Feb 20, 2020
@dweindl
Copy link
Member

dweindl commented Feb 28, 2020

See also #944

@dweindl
Copy link
Member

dweindl commented Mar 6, 2020

Implemented for PEtab test cases in tests/petab_test_suite. Might still be good to test for some more models...

@dweindl dweindl mentioned this issue Mar 7, 2020
Closed
@dweindl dweindl added ci enhancement and removed new Newly created labels Mar 30, 2020
@dweindl dweindl linked a pull request Apr 15, 2021 that will close this issue
Remove sllh computation from petab_objective.simulate_petab #1498
Merged
@dweindl dweindl removed a link to a pull request Apr 16, 2021
Remove sllh computation from petab_objective.simulate_petab #1498
Merged
@FFroehlich
Copy link
Member

https://github.com/wesselb/fdm might be relevant here

@dweindl
Copy link
Member

dweindl commented Aug 18, 2023

Done in #1548

@dweindl dweindl closed this as completed Aug 18, 2023
@dweindl dweindl mentioned this issue Oct 9, 2024
Merged
dweindl added a commit to dweindl/AMICI that referenced this issue Oct 9, 2024
@dweindl
CI: Re-enable benchmark model gradient checks / prettify gradient che…
b52340e 
…ck output

* Make the result more informative and more readable.
* Re-enable some test problems, use stricter tolerances for some at least AMICI-dev#967
* More problem-specific settings
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
ci enhancement
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@FFroehlich @dweindl @yannikschaelte