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

Reformatting, documentation, and slight API tweaks #1827

Merged
merged 10 commits into from
Apr 7, 2021

Conversation

benson31
Copy link
Collaborator

@benson31 benson31 commented Mar 12, 2021

The API tweaks are:

  • Made execution_context and termination_criteria polymorphically destructible.
  • Modified the API of training_algorithm to have get_type(), which returns a string description of the algorithm's type ("sgd", etc.), and get_name(), which allows users to assign an identifier to the algorithm.
  • Corrected the signature of training_algorithm::setup_models(...).
  • training_algorithms are now Cloneable.

The rest is documentation and formatting.

Copy link
Collaborator

@mrwyattii mrwyattii left a comment

Choose a reason for hiding this comment

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

LGTM!

NOTE: training_algorithm::get_type() is not entirely correct
here. It could cause problems if multiple instances of the same
callback are checkpointed at the same time, but currently we
don't do this. The issue with "get_name()" is that we might not
know the name of the training algorithm on restore (since the
training algos themselves are not checkpointed).
@bvanessen bvanessen merged commit 6c64f18 into LLNL:develop Apr 7, 2021
@benson31 benson31 deleted the training-algo-tweaks_test branch April 23, 2021 18:24
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants