Closed lsoucasse closed 1 month ago
Thanks for noticing this Laurent. Good to see the oscillations have gone away, although it's surprising given that the relative step tolerance is now higher/looser. We should look into what's causing them.
I guess maybe it's the prevent_warming
flag that damps the oscillations in this case, while when it's turned off the model is then free to oscillate around the F=0 state.
Need your approval again @nichollsh
I realized that the dummy integration test was not triggered because it was only a fixture of the plotting tests that were allowed to fail. With the current state of the repo, the dummy integration test is actually failing. I revert the dummy set up to a previous config which did not trigger spurious oscillation on the flux. I also switched the star model to use the Baraffe track.