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.

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

Add SOAP forcefields #19

Merged
merged 16 commits into from
Oct 18, 2023
Merged

Add SOAP forcefields #19

merged 16 commits into from
Oct 18, 2023

Conversation

WardLT
Copy link
Member

@WardLT WardLT commented Oct 5, 2023

Use SOAP, as implemented in DScribe, as features for a GPR regression model.

Still need to

  • Use GPU for training
  • Adjust hyperparameters
  • Build EnergyModel wrapper
  • Test for example problems

@WardLT WardLT marked this pull request as ready for review October 18, 2023 12:16
@WardLT
Copy link
Member Author

WardLT commented Oct 18, 2023

There's still more to do with SOAP, but let's merge what we have now. The forces predicted by SOAP are very poor compared to those from MBTR, and I suspect this has to do with how the model's trained or the choice of hyperparameters rather than a mistake in the software (see unit test comparing numerical to analytic forces).

Hoping to speed up solution times
Hoping to continue speeding solution
@WardLT WardLT merged commit e051bb4 into main Oct 18, 2023
0 of 2 checks passed
@WardLT WardLT deleted the soap_gpy branch December 25, 2023 17:24
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant