-
Notifications
You must be signed in to change notification settings - Fork 229
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.
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
Consistent Non-deterministic Behavior in Thermo (polycylic) and Kinetics (Peroxyl_Disproportionation) #2490
Comments
Probably a different bug but it's similar symptom:
at first I was surprised by the 10 orders of magnitude change in kinetics given that they used identical nodes, but then noticed it's because the thermo is different
it seems one matches oxirene and the other cyclopropene. |
This issue is being automatically marked as stale because it has not received any interaction in the last 90 days. Please leave a comment if this is still a relevant issue, otherwise it will automatically be closed in 30 days. |
I am broadening the scope of this issue since similar non-deterministic behavior has been detected in thermo estimations as well. The 'good' news is that the examples are reproducible (consistent). @xiaoruiDong has offered to investigate further (thanks!). |
I think the kinetics issue would resolve if we restructured Peroxyl_Disproportionation to operate like R_Recombination, where the labels on each reactant are the same, doing that would be a matter of doing a find-replace on labels in the training dictionary file, adapting the top group, regenerating the tree and then tweaking some logic in the database (should be able to search for mentions of "R_Recombination" to find the locations). |
@mjohnson541 let's catch up offline at some point about how to do this - we should document it for future devs as well. |
Bug Description
The regression tests are showing up differences between models, when no relevant changes to RMG have been made. The differences seem to be random
and in a different model
See my comment #2452 (comment)
and Matt's reply
#2452 (comment)
How To Reproduce
Run (or look at) the regression tests for a pull request that doesn't change RMG code (eg. it makes some other change). Two models built with identical code will have different results.
Expected Behavior
RMG results wouldn't be random.
Installation Information
Describe your installation method and system information.
The text was updated successfully, but these errors were encountered: