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

Set evaluate_every_epoch value based on number of train epochs completed #968

Closed
wants to merge 1 commit into from

Conversation

clarkdykang
Copy link
Contributor

Summary:

  • evaluate_every_n_epoch until having trained for a certain amount of epochs, after which we evaluate every epoch forcefully
  • Save time

Differential Revision: D68933995

@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D68933995

clarkdykang added a commit to clarkdykang/tnt that referenced this pull request Jan 31, 2025
…ted (pytorch#968)

Summary:

- `evaluate_every_n_epoch` until having trained for a certain amount of epochs, after which we evaluate every epoch forcefully
- Save total time by cutting down evaluations

Differential Revision: D68933995
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D68933995

…ted (pytorch#968)

Summary:

- Use `evaluate_every_n_epoch` until having trained for a certain amount of epochs, after which we evaluate every epoch forcefully
- Need to be able to set the `evaluate_every_n_epoch` field of the `PhaseState` class
- Save total time by cutting down evaluations, 8.3 hours -> 6.3 hours

Differential Revision: D68933995
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D68933995

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.

2 participants