This repository has been archived by the owner on Nov 3, 2023. It is now read-only.
[train] Fix bugs with loading validation impatience. #3713
Merged
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Outdated suggestions cannot be applied.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.
Patch description
Validation impatience has been loading incorrectly upon preemption for a bit.
The following two bugs are observed and fixed:
.checkpoint
was saved BEFORE the validation report had happened, meaning it would always save epochs/progress/etc with the latest values, but the "best validation score" of the PREVIOUS validation.impatience
key was not being set, and therefore always being loaded with 0 (the default).Testing steps
About to write unit tests.