Remove logspew from percentile es #3366
Closed
Conversation
This file contains 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
Summary: Playing whack-a-mode with some logs while developing tutorials. Logs are in an unacceptable state and take up more than 99% of the page real estate before this diff, producing multiple logs per progression call to `should_stop_trial_early` regardless of whether anything happened. We do not need this much visibility on whether or not we decide to kill a trial. In the tutorials we're working on (ex N6570044 we were seeing thousands of logs before this change, now we see a couple dozen). Changes: * No longer log if there havent been enough completed trials to make an ES decision * No longer log if no trials have met minimum observed progressions * No longer log the value of every trial before we make an ES decision (this happened twice) * No longer log at the beginning of ever ES decision that we are about to make an ES decision * No longer log when we consider stopping a trial and chose not to (only log if we actually decide it should stop) * [RFC] No longer log if an error happened during interpolation and we decide to use uninterpolated values instead (its not clear to me how important this error is -- fwiw in the tutorial we run into this very frequently -- if its a real actionable issue we need to fix it) Differential Revision: D69616469
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D69616469 |
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #3366 +/- ##
==========================================
- Coverage 95.73% 95.73% -0.01%
==========================================
Files 534 534
Lines 52580 52564 -16
==========================================
- Hits 50336 50320 -16
Misses 2244 2244 ☔ View full report in Codecov by Sentry. |
|
This pull request has been merged in 13ad726. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Summary:
Playing whack-a-mode with some logs while developing tutorials. Logs are in an unacceptable state and take up more than 99% of the page real estate before this diff, producing multiple logs per progression call to
should_stop_trial_earlyregardless of whether anything happened. We do not need this much visibility on whether or not we decide to kill a trial.In the tutorials we're working on (ex N6570044 we were seeing thousands of logs before this change, now we see a couple dozen).
Changes:
Differential Revision: D69616469