dEdLat contribution for stress calculations is added and Universal Potentials are updated #189
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…x and little cleanup in united tests
Codecov ReportAttention:
Additional details and impacted files@@ Coverage Diff @@
## main #189 +/- ##
==========================================
+ Coverage 98.00% 98.59% +0.59%
==========================================
Files 28 28
Lines 1901 1928 +27
==========================================
+ Hits 1863 1901 +38
+ Misses 38 27 -11 ☔ View full report in Codecov by Sentry. |
|
I'm checking out the source in VS Code, and I jumped in from this merged PR. Not sure if this is the right place to ask, but I've got a question about g.edata["pbc_offshift"] = torch.matmul(g.edata["pbc_offset"], lat[0])
g.ndata["pos"] = g.ndata["frac_coords"] @ lat[0]If so, then how does the subsequent code operate, such as the function dst_pos = g.ndata["pos"][g.edges()[1]] + g.edata["pbc_offshift"]
src_pos = g.ndata["pos"][g.edges()[0]]
bond_vec = dst_pos - src_pos
bond_dist = torch.norm(bond_vec, dim=1)Jiang You |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
I finally figured out that the stress contribution due to the change of lattice vectors in the Potential class is missing in the current implementation, now I have modified the implementations to add the strain (st) variable for lattices to include the gradient of potential energy with respect to lattice vectors for stress calculations. I also added unit tests for checking force and stress implementations with central finite difference methods and completed the retraining of both universal M3GNet MLIPs. The resulting pretrained universal MLIPs achieve better accuracy and stability in benchmark examples. Finally, I have checked all Jupyter Notebook examples, and everything works for the implementation.
Checklist
ruff.mypy.duecredit@due.dcitedecorators to reference relevant papers by DOI (example)Tip: Install
pre-commithooks to auto-check types and linting before every commit: