Add tip for logging evaluation metrics during regular evaluations #4367
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This PR adds a helpful comment in the DPO script explaining how to log and save evaluation metrics during regular evaluations using a custom callback. It also clarifies W&B behavior regarding metric aggregation. Related to issue huggingface#2602. Checklist: - [x] Added clear example for custom callback - [x] Clarified W&B aggregation behavior - [x] No code logic changed, only documentation tip
|
Hi team! This PR addresses the question raised in issue #2602 by adding a clear tip on how to log evaluation metrics during regular evaluations using a custom callback. |
|
thanks @cam1llynha. In my opinion, this belongs to the transformers documentation as it's common to all trainers |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR adds guidance in the DPO script on how to log and save evaluation metrics during regular evaluations, not only at the end of training.
The implementation includes an example of a custom callback (TrainerCallback) that allows logging intermediate metrics and clarifies how W&B (Weights & Biases) aggregates metrics across the evaluation dataset.
The goal is to provide clear reference for users who want to monitor metrics throughout the process without changing the core logic.
Related Issue:
Closes #2602