Repository navigation
Import batch-06; report agreement, not accuracy, for model-drafted labels - #31
Merged
Merged
Conversation
… labels All 60 batch-06 dev records are model drafts submitted unchanged. Labels after the pilot stay model-drafted, so new results are reported as agreement with the drafting model, and the release gates (judged on hand-labelled test records, of which there are 4) cannot be judged as declared. No target is changed.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
annotations/batches/batch-06.jsonl: 60 dev-pool records (natural sample), allmodel_draft_accepted, no annotation time recorded. No label-mix shift flagged against the earlier batches.docs/targets.md: status note and changelog entry. The gates are judged on hand-labelled test records, of which there are only 4, so they cannot be judged as declared. No target is changed.README.md,docs/evaluation.md,docs/model_card.md,docs/dataset_card.md: new results are reported as agreement with the drafting model, never as accuracy. Counts: 472 advisories annotated, 22 by hand (10 dev, 4 test).docs/annotation_plan.md,annotations/README.md: batch-06 recorded and the decision noted. The hand-labelling procedure is kept in case labelling resumes.Checks
ruff check,ruff format --check,pytest -m "not slow"pass; the committed-annotations test covers the new file.