Skip to content

fix single precision model compression error#1212

Merged
wanghan-iapcm merged 1 commit intodeepmodeling:develfrom
denghuilu:fix-single-compress
Oct 15, 2021
Merged

fix single precision model compression error#1212
wanghan-iapcm merged 1 commit intodeepmodeling:develfrom
denghuilu:fix-single-compress

Conversation

@denghuilu
Copy link
Member

There's an error when using the single precision model compression:

input ‘em_x’ of tabulatefusion op has type float32 that does not match type float64 of augment ‘table’

This fault is mainly caused by the default type setting of the custom op TabulateFusion.

This PR should fix this problem.

@denghuilu denghuilu requested a review from njzjz October 14, 2021 05:33
@codecov-commenter
Copy link

codecov-commenter commented Oct 14, 2021

Codecov Report

Merging #1212 (bc24ad4) into devel (35d333e) will decrease coverage by 0.06%.
The diff coverage is 100.00%.

Impacted file tree graph

@@            Coverage Diff             @@
##            devel    #1212      +/-   ##
==========================================
- Coverage   76.04%   75.98%   -0.07%     
==========================================
  Files          91       86       -5     
  Lines        7260     7245      -15     
==========================================
- Hits         5521     5505      -16     
- Misses       1739     1740       +1     
Impacted Files Coverage Δ
deepmd/descriptor/se_a.py 96.25% <100.00%> (-0.02%) ⬇️
deepmd/common.py 80.00% <0.00%> (-4.29%) ⬇️
source/api_cc/include/DataModifier.h
source/lib/include/SimulationRegion.h
source/api_cc/include/DeepPot.h
source/api_cc/include/DeepTensor.h
source/lib/include/neighbor_list.h

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 35d333e...bc24ad4. Read the comment docs.

@wanghan-iapcm wanghan-iapcm merged commit 4bc3b27 into deepmodeling:devel Oct 15, 2021
njzjz added a commit to njzjz/deepmd-kit that referenced this pull request Sep 21, 2023
The `from_struct["atom_types"]` used to be a list but was changed to
numpy array lately, thereby introducing a bug in `make_supercell_abacus`
function.
Also, the key `types` is changed to `atom_types` outside the function,
but not changed in the function.
 This commit fix the two bugs.

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
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.

4 participants