-
Notifications
You must be signed in to change notification settings - Fork 423
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
Adding logging and force flushing for run events #2703
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
QQ: are these dditional logs generally helpful for future debugging or are they just useful for debugging the issues we are seeing in the current run?
I think if we get logging to occur consistently, then we can remove these. |
@jjanezhang should we not merge this change then and simply set the integration in the RunConfig when debugging the issue?
Alternatively if you think this improves the mosaicml_logger generally then we should merge it. |
@j316chuck not a lot of harm in adding debug logs :) |
I can't add integrations to a finetune yaml so the best I can do is submit a normal run with a ft generated yaml and then add the integrations. If this is really a concern I can just test with the later, but I would prefer to be able to check the logs from a finetune run. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Approved with one minor comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM -- agree with Daniel to make sure logs look OK and not spammy. Spammy logs = no one reads anything
Adding logging and force flushing for run events
We are seeing run events inconsistently sent to run metadata for finetuning runs. Adding logs to help debug and force flushing metadata when run event related metadata is being written to the run.
Testing