Repository navigation
Re-judge past validator fixes with today's validator (to a CSV, no writes) - #908
Merged
Merged
Conversation
… a CSV A wrong FIX never heals: the corrected meaning is VALID and never looked at again. Batch API for the first question; the validator's own retries live. Writes nothing to the database. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
ArchLens - No architecturally relevant changes to the existing views |
…e rest later don't pay twice Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Roadmap item 2: re-check past validator fixes.
A wrong INVALID heals itself (scheduling and the nightly job re-judge anything not VALID), but a wrong FIX doesn't: the learner moves to a corrected meaning marked VALID, which is never looked at again. "vil sige = mean" (13 Aug 2026) is one.
tools/recheck_validator_fixes.pyasks today's validator about each corrected meaning in the sentence the fix was made in, and writes the ones it judges differently to a CSV, most active words first. It writes nothing to the database.validate_and_classifyas production.--checked(a file of validation_log ids, defaultvalidator_fix_rechecked.txt) and is skipped next run, so the top 300 now and the rest next month don't pay twice. Keep the file between runs.--dry-runcounts and estimates;--days,--limitnarrow it;--batch-idresumes after an interruption.Checked: the query on prod (read-only); the batch-then-live flow with fakes; one real request with the batch's parameters ("vil sige" = "mean" → FIX "means").
🤖 Generated with Claude Code