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

Load a checkpoint file from wandb directly into a potential. #311

Merged
merged 20 commits into from
Nov 14, 2024

Conversation

chrisiacovella
Copy link
Member

@chrisiacovella chrisiacovella commented Nov 6, 2024

Pull Request Summary

It would be convenient to be able to setup a potential directly from a checkpoint file on weights and biases. This PR implements such functionality.

Key changes

Notable points that this PR has either accomplished or will accomplish.

  • This PR adds a function to the NeuralNetworkPotentialFactory that will download the file and call the load_inference_model_from_checkpoint function, returning a potential.

Questions

  • I'm not sure the best way to test this since it will require credentials to access Wandb. For now I've marked the test to skip on GitHub. Obviously won't work if someone is not part of the project, but that is fine. This is just for convenience of us sharing models during development and testing.

Associated Issue(s)

Pull Request Checklist

  • Issue(s) raised/addressed and linked
  • Includes appropriate unit test(s)
  • Appropriate docstring(s) added/updated
  • Appropriate .rst doc file(s) added/updated
  • PR is ready for review

@codecov-commenter
Copy link

codecov-commenter commented Nov 6, 2024

Codecov Report

Attention: Patch coverage is 23.07692% with 10 lines in your changes missing coverage. Please review.

Project coverage is 34.18%. Comparing base (dbae94c) to head (ee7e6cf).

❗ There is a different number of reports uploaded between BASE (dbae94c) and HEAD (ee7e6cf). Click for more details.

HEAD has 7 uploads less than BASE
Flag BASE (dbae94c) HEAD (ee7e6cf)
unittests 8 1
Additional details and impacted files

… load_inference_model_from_checkpoint now accepts an optional argument "only_unique_pairs" to set this when reading a checkpoint file, for any models trained prior to PR choderalab#299 (openmm integration).
@chrisiacovella
Copy link
Member Author

I'm just going to merge this because it doesn't make any really substantial changes.

@chrisiacovella chrisiacovella merged commit 80176df into choderalab:main Nov 14, 2024
10 checks passed
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.

2 participants