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

Implement fixed composition weights provided by the user #129

Closed
wants to merge 2 commits into from

Conversation

SanggyuChong
Copy link
Contributor

@SanggyuChong SanggyuChong commented Mar 1, 2024

Hello,

This is an attempt at implementing the feature where users are able to define the composition weights used in training the SOAP-BPNN model.

  • User, if he/she desires, can specify a dictionary of dictionaries in the training section of the YAML, where the key to the sub-dictionary is the target, and sub-dictionary consists of species atomic numbers and their fixed composition weights.
  • For a behavior where the use of composition weights is to be turned off, user can use this feature and set all weights of all species to 0.
  • Checks if values are provided for all species present in the training dataset(s).

Resolves #108


📚 Documentation preview 📚: https://metatensor-models--129.org.readthedocs.build/en/129/

@SanggyuChong SanggyuChong added the SOAP BPNN SOAP BPNN experimental architecture label Mar 1, 2024
@SanggyuChong SanggyuChong deleted the update-yaml branch March 1, 2024 17:16
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
SOAP BPNN SOAP BPNN experimental architecture
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Option to explicitly set isolated atom energies
1 participant