fix: package complete review validators
This commit is contained in:
@@ -16,6 +16,8 @@
|
|||||||
- Added a symmetric fail-closed false-negative decision validator and readiness
|
- Added a symmetric fail-closed false-negative decision validator and readiness
|
||||||
compilation gate. It exports only explicit confirmed misses and rejects
|
compilation gate. It exports only explicit confirmed misses and rejects
|
||||||
incomplete, mismatched or invalid decision sets.
|
incomplete, mismatched or invalid decision sets.
|
||||||
|
- Packaged the validator in the all-in-one Unraid image and covered that runtime
|
||||||
|
file contract with the existing Docker configuration regression.
|
||||||
- Audited the confirmed evidence against the active tile corpus and holdouts.
|
- Audited the confirmed evidence against the active tile corpus and holdouts.
|
||||||
Geel/Herentals evidence already belongs to the training source and Turnhout
|
Geel/Herentals evidence already belongs to the training source and Turnhout
|
||||||
remains excluded, leaving zero novel leakage-free labels. No model was
|
remains excluded, leaving zero novel leakage-free labels. No model was
|
||||||
|
|||||||
@@ -82,7 +82,9 @@ def test_all_in_one_dockerfile_copies_operator_scripts_for_runtime_use() -> None
|
|||||||
"audit_detection_false_negative_evidence.py",
|
"audit_detection_false_negative_evidence.py",
|
||||||
"audit_detection_false_positive_evidence.py",
|
"audit_detection_false_positive_evidence.py",
|
||||||
"render_detection_false_positive_review_contact_sheets.py",
|
"render_detection_false_positive_review_contact_sheets.py",
|
||||||
|
"render_detection_false_negative_review_contact_sheets.py",
|
||||||
"validate_detection_false_positive_review_decisions.py",
|
"validate_detection_false_positive_review_decisions.py",
|
||||||
|
"validate_detection_false_negative_review_decisions.py",
|
||||||
"run_operator_hard_negative_detection_matrix.sh",
|
"run_operator_hard_negative_detection_matrix.sh",
|
||||||
"run_background_corpus_split_matrix.sh",
|
"run_background_corpus_split_matrix.sh",
|
||||||
"build_background_corpus_split_report.py",
|
"build_background_corpus_split_report.py",
|
||||||
|
|||||||
@@ -98,6 +98,7 @@ COPY scripts/audit_detection_false_positive_evidence.py /app/scripts/audit_detec
|
|||||||
COPY scripts/render_detection_false_positive_review_contact_sheets.py /app/scripts/render_detection_false_positive_review_contact_sheets.py
|
COPY scripts/render_detection_false_positive_review_contact_sheets.py /app/scripts/render_detection_false_positive_review_contact_sheets.py
|
||||||
COPY scripts/render_detection_false_negative_review_contact_sheets.py /app/scripts/render_detection_false_negative_review_contact_sheets.py
|
COPY scripts/render_detection_false_negative_review_contact_sheets.py /app/scripts/render_detection_false_negative_review_contact_sheets.py
|
||||||
COPY scripts/validate_detection_false_positive_review_decisions.py /app/scripts/validate_detection_false_positive_review_decisions.py
|
COPY scripts/validate_detection_false_positive_review_decisions.py /app/scripts/validate_detection_false_positive_review_decisions.py
|
||||||
|
COPY scripts/validate_detection_false_negative_review_decisions.py /app/scripts/validate_detection_false_negative_review_decisions.py
|
||||||
COPY scripts/run_operator_hard_negative_detection_matrix.sh /app/scripts/run_operator_hard_negative_detection_matrix.sh
|
COPY scripts/run_operator_hard_negative_detection_matrix.sh /app/scripts/run_operator_hard_negative_detection_matrix.sh
|
||||||
COPY scripts/run_background_corpus_split_matrix.sh /app/scripts/run_background_corpus_split_matrix.sh
|
COPY scripts/run_background_corpus_split_matrix.sh /app/scripts/run_background_corpus_split_matrix.sh
|
||||||
COPY scripts/build_background_corpus_split_report.py /app/scripts/build_background_corpus_split_report.py
|
COPY scripts/build_background_corpus_split_report.py /app/scripts/build_background_corpus_split_report.py
|
||||||
|
|||||||
@@ -8205,6 +8205,8 @@ Implemented:
|
|||||||
and 12 uncertain/obscured cases.
|
and 12 uncertain/obscured cases.
|
||||||
- Added `validate_detection_false_negative_review_decisions.py`, mirroring the
|
- Added `validate_detection_false_negative_review_decisions.py`, mirroring the
|
||||||
existing FP safety contract and exporting only explicit confirmed misses.
|
existing FP safety contract and exporting only explicit confirmed misses.
|
||||||
|
- Included the symmetric validator in the all-in-one image and its packaging
|
||||||
|
regression list after the first live build exposed the missing COPY entry.
|
||||||
- Audited confirmed evidence against the active training split. Geel and
|
- Audited confirmed evidence against the active training split. Geel and
|
||||||
Herentals are existing training sources; Turnhout is an excluded operational
|
Herentals are existing training sources; Turnhout is an excluded operational
|
||||||
holdout. The review therefore provides zero novel leakage-free labels and a
|
holdout. The review therefore provides zero novel leakage-free labels and a
|
||||||
|
|||||||
Reference in New Issue
Block a user