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

assign energy shift stats if atomic energies are assigned #1477

Merged
merged 5 commits into from
Feb 16, 2022

Conversation

njzjz
Copy link
Member

@njzjz njzjz commented Feb 15, 2022

Atomic energies stats may be incorrect. Here, we use assigned atomic energies instead.
I am training a model against multiple systems, where some atoms should not contribute energies. Before this commit, the loss of my model could not decrease. After this commit, it decreased quickly.

Atomic energies stats may be incorrect. Here, we use assigned atomic energies instead.
I am training a model against multiple systems, where some atoms should not contribute energies. Before this commit, the loss of my model could not decrease. After this commit, it decreased quickly.
@codecov-commenter
Copy link

codecov-commenter commented Feb 15, 2022

Codecov Report

Merging #1477 (352a861) into devel (82c787d) will increase coverage by 0.33%.
The diff coverage is 100.00%.

Impacted file tree graph

@@            Coverage Diff             @@
##            devel    #1477      +/-   ##
==========================================
+ Coverage   75.68%   76.01%   +0.33%     
==========================================
  Files          92       92              
  Lines        7671     7681      +10     
==========================================
+ Hits         5806     5839      +33     
+ Misses       1865     1842      -23     
Impacted Files Coverage Δ
deepmd/fit/ener.py 92.54% <100.00%> (+1.63%) ⬆️
deepmd/descriptor/se_a.py 94.82% <0.00%> (+0.64%) ⬆️
deepmd/train/trainer.py 74.53% <0.00%> (+2.08%) ⬆️
deepmd/utils/network.py 92.63% <0.00%> (+9.83%) ⬆️

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 82c787d...352a861. Read the comment docs.

Copy link
Collaborator

@wanghan-iapcm wanghan-iapcm left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could you please add a UT for your update?
For example you may provide 1 atom_ener for a three component system, and see if the computed energy_shift exactly use the assigned atom_ener and fits the other two element from the data.
(I hope I understood your PR correctly..)

deepmd/fit/ener.py Outdated Show resolved Hide resolved
@njzjz njzjz marked this pull request as draft February 15, 2022 16:52
@njzjz njzjz marked this pull request as ready for review February 15, 2022 19:21
@wanghan-iapcm wanghan-iapcm merged commit b26ef18 into deepmodeling:devel Feb 16, 2022
@njzjz njzjz deleted the atom_ener_stat branch February 17, 2022 05:03
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.

None yet

3 participants