Fix: bug of training force and stress with DeePKS #3444
Merged
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
| for (int iw2l = 0; iw2l < col_indexes.size(); ++iw2l) | ||
| { | ||
| std::vector<double> nlm1 = this->nlm_save_k[iat][key_1][row_indexes[iw1l]][0]; |
There was a problem hiding this comment.
Plz add a UnitTest to trace similar error.
There was a problem hiding this comment.
I have added check lists in deepks test cases for both multi-k and gamma-only cases.
The UnitTest will be implemented in future PRs.
|
I am still not convinced by this pr, will raise an issue about this |
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Outdated suggestions cannot be applied.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.
When user training DeePKS model with ABACUS, force and stress labels contain two gradient files named "grad_vx.npy" and "grad_vepsl.npy".
These two files are needed by DeePKS-kit but don't affect the self-consistent computation with loading DeePKS model in ABACUS.
Before #3444 , when "cal_stress = 1", the line of " https://github.com/deepmodeling/abacus-develop/blob/v3.5.0/source/module_hamilt_lcao/module_deepks/LCAO_deepks_pdm.cpp#L869 " conflict with the line of "https://github.com/deepmodeling/abacus-develop/blob/v3.5.0/source/module_hamilt_lcao/module_deepks/LCAO_deepks_pdm.cpp#L808" , which will cause severe bug that "grad_vx.npy" and "grad_vepsl.npy" are wrong and convergence of DeePKS model training won't be achieved.
This bug will be fixed after #3444 , I also add checklists in deepks test cases for protecting these code.