Compute validation metrics at first step #1508
Merged
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
meta-cla
bot
added
the
CLA Signed
tianyu-l
reviewed
Aug 1, 2025
Contributor
tianyu-l
left a comment
tianyu-l
left a comment
There was a problem hiding this comment.
I thought about this before. I think most desirable thing to do is do validation before step 1 trained, which would cause bigger change to train.py. The code change here would do it after step 1 trained. Is this what you want? (I'm OK with this.)
Contributor
Author
|
@tianyu-l I agree that validation before the first step might in principle be desirable, but I don't think it's worth a larger change to train.py. In practice, I think running validation after the first step should almost always be sufficient. |
tianyu-l
approved these changes
Aug 1, 2025
bentherien
pushed a commit
to bentherien/torchtitan_
that referenced
this pull request
Aug 5, 2025
Currently, the first time validation metrics are computed is when `step == job_config.validation.freq`. I think it is preferable to always compute them for the first step as well.
joellidin
pushed a commit
to one-covenant/torchtitan
that referenced
this pull request
Aug 8, 2025
Currently, the first time validation metrics are computed is when `step == job_config.validation.freq`. I think it is preferable to always compute them for the first step as well.
joellidin
pushed a commit
to one-covenant/torchtitan
that referenced
this pull request
Aug 8, 2025
Currently, the first time validation metrics are computed is when `step == job_config.validation.freq`. I think it is preferable to always compute them for the first step as well.
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.
Currently, the first time validation metrics are computed is when
step == job_config.validation.freq. I think it is preferable to always compute them for the first step as well.