-
Notifications
You must be signed in to change notification settings - Fork 227
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
ArrheniusBM fitting procedure #2264
Conversation
998126f
to
fb42d78
Compare
I modified the ArrBM
On the main branch, the test fails with this error
The test passes on PR branch |
I've just rebased this onto the official main. I'm not sure why we'd ever have a negative |
I have some similar fixes on the electrochem branch. After discussion we decided to simply average reactions instead of using the BM fit when E0<0. |
We decided that the BM fit shouldn't be expected to be valid for submerged barrier reactions. |
Is it these ones, Matt? #2277 |
We previously did not account for E0 < 0. Here, for E0 < 0, we set Ea to E0 for exothermic rxns and dHrxn for endothermic. We also incorporate aspects of Reaction `fix_barrier_height` method to set Ea to 0 for exothermic rxns and Ea < 0, and set Ea to dHrxn if Ea < dHrxn for endothermic rxns.
Use the `get_activation_energy` method in the objective function. For intial guess of params, use average of A and n, and use BEP for guess of Ea.
Test now compares the fitted rate to the rate it was trained on to make sure they agree.
Codecov Report
@@ Coverage Diff @@
## main #2264 +/- ##
=======================================
Coverage 48.14% 48.14%
=======================================
Files 110 110
Lines 30629 30629
Branches 7989 7989
=======================================
Hits 14747 14747
Misses 14353 14353
Partials 1529 1529 📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
Closing in favor of #2316 for the time being, but if that doesn't pan out we can come back to this one. |
Motivation or Problem
dHrxn > 4 * w0/10
ordHrxn < -4 * w0/10
Description of Changes
get_activation_energy
to handle E0 < 0get_activation_energy
method when getting training rates and fitting paramsTesting
I used this to regenerate rules for a bunch of families. It doesn't change too much, except for rules trained on single rxn (if rxn is very exothermic with high Ea)
Reviewer Tips
get_activation_energy
method and fitting to single training reactions.