Implementation of smooth D4S#264
Conversation
awvwgk
left a comment
awvwgk
left a comment
There was a problem hiding this comment.
This seems to be API breaking changes in dftd4, so we need to bump it to version 4 if the new model cannot be implemented in an API compatible way.
|
Yes, I think this will break the API. Because the Exponents wf are now pairwise defined in D4S there is no way of writing gwvec with only the information of one atom (i.e. it needs a third dimension). I don't think this can be avoided. |
|
I added a small fix to the CI and wheels so that the tests are executed. The CI should generally be updated (similar to tblite). I will look into it but in a separate PR. |
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
4 participants
Implementation of the smoothed D4S version using pairwise exponents in the averaging of the CN (10.26434/chemrxiv-2024-31x2z). The changes include:
Still to do: