feat(annotation): panel-status core engine by henrycgbaker · Pull Request #248 · bertelsmannstift/pragmata

henrycgbaker · 2026-06-01T11:39:42Z

Stack (6 PRs): #246 → #247 → #268 → #248 → #269 → #249

This PR (4/6): Core engine for live per-panel completeness + all-task progress, computed directly against Argilla (no export round-trip). Exposed via CLI/API in #269. Stacked on #268.

All tasks: walks every dataset and computes record progress (total / completed) across retrieval / grounding / generation.
Retrieval panels: the retrieval row also carries panel-completeness (panel_complete = all K chunks have ≥1 submitted, overlap_satisfied = each chunk meets its min_submitted, integrity vs n_retrieved_chunks).
Config-free + multi-domain: selects datasets by name-prefix (no AnnotationSettings), covering every workspace in one pass. Panels keyed by (workspace, record_uuid). min_submitted read live from each dataset's Argilla settings.

The CLI/API surface (pragmata annotation status) ships in #269. The --tag-partial-panels write ships in #249.

ddimmery

The naming of ProgressReport / StatusReport etc could be made more clear to distinguish the hierarchy.

ddimmery · 2026-07-04T10:46:34Z

+            )
+            n_integrity_warnings += 1
+        if facts.k_metadata == 0:
+            n_panels_unknown_k += 1


Is this worth exposing in the StatusReport?

ddimmery · 2026-07-06T07:20:42Z

+        integrity_ok = facts.k_metadata == 0 or facts.k_metadata == facts.k_records
+        if not integrity_ok:
+            logger.warning(
+                "record_uuid=%s: integrity warning - %d records but n_retrieved_chunks metadata=%d",


shouldn't we also print out ws_name?

ddimmery · 2026-07-06T12:48:14Z

+    n_overlap_satisfied: int
+    n_integrity_warnings: int
+    n_orphans_skipped: int
+    progress: "ProgressReport | None" = None


Is there a reason for this to be optional?

**Stack (6 PRs):** #246 → #247 → #268 → #248 → #269 → #249 **Chain goal:** Persist K (number of retrieved chunks per query) on retrieval records, surface it as panel-completeness columns + sidecar at export time, ship a live `annotation status` CLI on top, and add an opt-in `--tag-partial-panels` advisory write so annotators can filter partially-done panels in the Argilla UI. **This PR (1/6):** Foundation. Stamps `n_retrieved_chunks` on retrieval record metadata at import and surfaces it on `RetrievalAnnotation`. Ships nothing user-visible on its own — the export columns (#247/#268) and live status CLI (#248/#269) consume it. --- ## Scope ### Record building - `core/annotation/record_builder.py`: add `"n_retrieved_chunks": len(pair.chunks)` to every chunk-record's metadata dict. ### Task definition - `core/annotation/argilla_task_definitions.py`: declare `IntegerMetadataProperty("n_retrieved_chunks", min=1, visible_for_annotators=False)` on retrieval task settings. ### Schema - `core/schemas/annotation_export.py`: `RetrievalAnnotation.n_retrieved_chunks: int` field (mirrors the existing `chunk_rank: int`). ### Export readback - `core/annotation/export_fetcher.py`: `_build_row` retrieval branch reads `metadata.get("n_retrieved_chunks", 0)` (defensive default for records imported before this PR). ### Tests - New `test_n_retrieved_chunks_stamped_on_every_chunk_record` in `test_import.py` locks the persist contract. - Metadata-name list in `test_argilla_task_definitions.py` updated. - Fixture refresh across `test_annotation_export.py`, `test_export_runner.py`, `test_export_constraint_checks.py`, `test_iaa_runner.py`, `test_export_api.py`. ## Why K varies per query (~15% K<5, 61% K=5, 24% K>5; the retriever is threshold-based, not top-K), so the export side can't infer K from `chunk_rank` or from record-counting alone — it needs a persisted SSOT to drive true top-K metrics in the consumer pipeline (precision@K, recall@K, NDCG@K, MRR@K). End-to-end mirror of the existing `chunk_rank` field: stamped at import, declared on the task, modelled on `RetrievalAnnotation`, read back by the fetcher. A one-off backfill script for records imported before this PR (in-flight batches) lives outside the supported CLI/API surface — retrospective ops migration, not packaged functionality. ## Base `feat/querygen-resumable-stack` — kitchen-sink integration branch carrying the full dep surface this stack assumes (`Locale`, `atomic_write_text`, `LogicalConstraint`, `WIDGET_FIELD_PLACEHOLDERS`, the `export_constraint_checks` rename). Rebases cleanly to `main` once those upstream stacks land. ## Test plan - [x] `uv run python -m pytest tests/unit` (863 passing) ---

**Stack (6 PRs):** #246 (merged) → #247 → #268 → #248 → #269 → #249 **Chain goal:** Persist K (number of retrieved chunks per query) on retrieval records, surface it as panel-completeness columns + sidecar at export time, ship a live `annotation status` CLI on top, and add an opt-in `--tag-partial-panels` advisory write so annotators can filter partially-done panels in the Argilla UI. **This PR (2/6):** The standalone, export-agnostic panel-completeness calculator, plus the shared retrieval-walk primitive it and #268 both consume. Split out of what was originally a single ~1000-line PR; wiring completeness into the retrieval export path ships separately in #268. --- ## Scope ### Completeness aggregator - New `core/annotation/completeness.py`: groups retrieval snapshots by `record_uuid`, distinct-by-`chunk_id`, derives per-panel facts, emits `CompletenessReport` (`by_uuid` + `summary`). Aggregation is split into a typed `_PanelAccumulator` plus `_aggregate_snapshots` (grouping) / `_summarize_groups` (transform) passes. ### Shared retrieval walk - `core/annotation/export_fetcher.py`: `RetrievalRecordSnapshot` + `walk_retrieval_records` — one Argilla scroll, no status filter (so records lacking terminal responses are still seen for the integrity check), feeding both this module's completeness pass and #268's row emission. Plus `resolve_task_purposes`, a topology lookup extracted from `fetch_task` (no behaviour change). ### Schema - `core/schemas/annotation_export.py`: `CompletenessSummary` + `KBucketStat` models for the export sidecar (populated by #268). ### Tests - New `test_completeness.py`: happy path, distinct-by-chunk_id, orphan exclusion, integrity warnings (records-vs-K + mixed-backfill within panel), all-unknown-K warning, K-bucket cross-tab + per-K histogram, STRICT `panel_complete` (discards do NOT count), `n_discarded_chunks` subset of `n_annotated_chunks`, calibration walked when topology declares it. ## Why STRICT `panel_complete` The computed `panel_complete: bool` is STRICT: True iff every K chunk in the panel has at least one **submitted** response. Discarded responses are abstentions, not judgements (pragmata's `DiscardReason` enum is refusal-only: `INVALID_OR_UNREALISTIC` / `UNCLEAR` / `OUTSIDE_REVIEWER_EXPERTISE`), so treating them as covered would feed unjudged chunks into NDCG@K / precision@K denominators as 0-relevance labels. The bool is derivable from the count columns, so a consumer wanting the permissive policy computes `n_annotated_chunks == n_retrieved_chunks` in one line. The retriever is threshold-based, so dropping partial panels biases metrics toward small K (MNAR); this module is the shared calculation both the export path (#268) and the live status path (#248) build on — one aggregator, two consumers. ## Test plan - [x] `uv run python -m pytest tests/unit` (1040 passing)

…ry) (#268) **Stack (6 PRs):** #246 → #247 → #268 → #248 → #269 → #249 **This PR (3/6):** Split out of #247 (was 1001 lines / XL). Wires the panel-completeness calculator (#247) into the retrieval export path. Stacked on #247. --- ## Scope ### Schema - `core/schemas/annotation_export.py`: - `RetrievalExportRow` += `panel_complete: bool`, `n_annotated_chunks: int`, `n_submitted_chunks: int`, `n_discarded_chunks: int`, `n_records_seen: int`. All defaulted; appended at the tail so existing column order is preserved. - New `KBucketStat` (per-K-bucket panel counts) and `CompletenessSummary` (aggregates + `by_k_bucket` cross-tab + full per-K histogram) Pydantic models. - `AnnotationExportMeta` += `completeness_summary: CompletenessSummary | None` and `completeness_status: Literal["ok","failed","not_requested"]`. ### Single-walk retrieval pipeline - `core/annotation/export_fetcher.py`: new `walk_retrieval_records` (one Argilla scroll per dataset, no status filter — sees discards-only records too so the integrity check has the full record set) + `RetrievalRecordSnapshot` dataclass + `fetch_retrieval_from_records` (status filter applied in Python at typed-row construction). - `core/annotation/export_runner.py`: when `Task.RETRIEVAL in tasks`, walks once and feeds both the typed-row projection and the completeness aggregator. `compute_completeness` failures degrade the sidecar (sets `completeness_status="failed"`) without losing the already-fetched export. ### API - `annotation/__init__.py`: lazy re-exports for `CompletenessSummary`, `KBucketStat`. ### Tests - `test_export_runner.py`: `write_export_csv` stamps completeness columns by `record_uuid`; sidecar carries `completeness_summary`; `completeness_status` ok/failed/not_requested set explicitly. - `test_annotation_export.py`: `RetrievalExportRow` field-order locked. ## Why The retriever is threshold-based, so dropping partial panels biases the metrics toward small K (MNAR). Surfacing both per-row predicate columns AND a bucketed sidecar aggregate lets the eval pipeline either filter complete panels per-K or apply condensed-list scoring without re-querying the data layer. The single-walk shape (one retrieval scroll feeds both fetch_task and compute_completeness) costs one extra in-memory pass over already-fetched records but spares the Argilla server a second full scroll per export — meaningful on the in-flight batches (~hundreds of panels). The STRICT `panel_complete` policy question (which shape to ship — this PR assumes (A) from #247) is discussed on #247, since that's where the predicate is actually computed; this PR just surfaces it as a column. ## Test plan - [ ] `uv run python -m pytest tests/unit/core/annotation/test_export_runner.py tests/unit/core/schemas/test_annotation_export.py`

Computes live per-panel completeness across every retrieval dataset in one pass: walks all datasets by name-prefix (no AnnotationSettings, so it's config-free and multi-domain), reads each dataset's min_submitted live from its Argilla distribution settings, and keys panels by (workspace, record_uuid) so sibling panels across tasks are never fused. Two distinct notions of complete are computed side by side so operators don't conflate them: - panel_complete (metric-facing, STRICT): every K chunk has at least one SUBMITTED response. - overlap_satisfied (operational): every chunk's submitted-response count is >= its per-purpose Argilla min_submitted threshold (1 prod / 3 cal). A panel can be metric-complete but overlap-unsatisfied, or vice versa. K is computed live by counting distinct chunk-records per record_uuid and cross-checked against the n_retrieved_chunks metadata for integrity; mismatches log a warning and surface in the report. compute_task_progress adds an all-task record-progress summary (total/completed across retrieval/grounding/generation) via dataset.progress(), aggregated grand/by-task/by-workspace/by-dataset. Orphan records (empty record_uuid) are excluded from panel grouping and reported separately.

…249) **Stack (6 PRs):** #246 → #247 → #268 → #248 → #269 → #249 **This PR (6/6):** Adds the advisory `needs_completion` tag write onto `annotation status`. - `status --tag-partial-panels`: stamps `needs_completion` on the **unresolved chunks of PARTIAL panels** (`0 < submitted < K`) and clears stale tags, sharing the single status walk. Annotators filter `needs_completion=true` in the Argilla UI to focus on the records that will complete a partially-done panel. - **Partial-only gate** (not "any incomplete panel"): a fully-unstarted panel is never tagged, so the tag stays a selective mop-up signal instead of ≈ everything pending. - **Multi-domain + prod/cal-split aware**: a panel whose chunks are split across a workspace's production + calibration datasets (per-item calibration) is evaluated as one unit; writes route to each chunk's owning dataset. Reuses `metadata_ops` for the replace-safe upsert. Stacked on #269. Also re-exports `TagResult` from the `pragmata.annotation` facade: `StatusReport.tag_result` is a public field of this type, but the facade export was pre-declared two PRs ago (before `panel_status.py` existed), dropped once that PR's version of the module still didn't define the class, and never re-added once this PR added it for real.

github-actions Bot added feature Adds or expands user-facing functionality size: L 500-999 LOC labels Jun 1, 2026

This was referenced Jun 1, 2026

feat(annotation): persist n_retrieved_chunks on retrieval records #246

Merged

feat(annotation): panel completeness calculator #247

Merged

feat(annotation): tag partial panels via status --tag-partial-panels #249

Merged

henrycgbaker changed the title ~~feat(annotation): add 'annotation status' command for live panel completeness~~ feat(annotation): live panel-status command (config-free, multi-domain) Jul 1, 2026

henrycgbaker changed the title ~~feat(annotation): live panel-status command (config-free, multi-domain)~~ feat(annotation): live annotation status (all-task progress + retrieval panels) Jul 1, 2026

henrycgbaker force-pushed the feat/annotation-retrieval-completeness/02-export-completeness branch from 42bd0be to 2791caa Compare July 1, 2026 16:37

henrycgbaker force-pushed the feat/annotation-retrieval-completeness/03-status-command branch from 481a9fd to b8f430b Compare July 1, 2026 16:37

github-actions Bot added size: XL 1000+ LOC and removed size: L 500-999 LOC labels Jul 1, 2026

henrycgbaker mentioned this pull request Jul 1, 2026

feat(annotation): status API + CLI command #269

Merged

1 task

henrycgbaker changed the base branch from feat/annotation-retrieval-completeness/02-export-completeness to feat/annotation-retrieval-completeness/02b-wire-export-completeness July 1, 2026 16:38

henrycgbaker mentioned this pull request Jul 1, 2026

feat(annotation): wire completeness into export (columns + meta summary) #268

Merged

1 task

henrycgbaker changed the title ~~feat(annotation): live annotation status (all-task progress + retrieval panels)~~ feat(annotation): panel-status core engine Jul 1, 2026

henrycgbaker force-pushed the feat/annotation-retrieval-completeness/03-status-command branch from b8f430b to c80a597 Compare July 3, 2026 13:34

henrycgbaker force-pushed the feat/annotation-retrieval-completeness/02b-wire-export-completeness branch from fd0eb3d to 2d18117 Compare July 3, 2026 13:34

github-actions Bot added size: L 500-999 LOC and removed size: XL 1000+ LOC labels Jul 3, 2026

henrycgbaker requested a review from ddimmery July 3, 2026 13:46

henrycgbaker marked this pull request as ready for review July 3, 2026 13:46

saschagobel added the annotation Changes affecting the annotation tool label Jul 4, 2026

ddimmery approved these changes Jul 6, 2026

View reviewed changes

henrycgbaker force-pushed the feat/annotation-retrieval-completeness/02b-wire-export-completeness branch from 2d18117 to de3f8ff Compare July 7, 2026 07:46

Base automatically changed from feat/annotation-retrieval-completeness/02b-wire-export-completeness to main July 7, 2026 08:43

henrycgbaker force-pushed the feat/annotation-retrieval-completeness/03-status-command branch from c80a597 to 7597597 Compare July 7, 2026 11:45

henrycgbaker force-pushed the feat/annotation-retrieval-completeness/03-status-command branch from 7597597 to 751883c Compare July 7, 2026 12:25

henrycgbaker merged commit 9f4b5b2 into main Jul 7, 2026
4 checks passed

henrycgbaker deleted the feat/annotation-retrieval-completeness/03-status-command branch July 7, 2026 12:27

henrycgbaker restored the feat/annotation-retrieval-completeness/03-status-command branch July 7, 2026 12:29

henrycgbaker deleted the feat/annotation-retrieval-completeness/03-status-command branch July 7, 2026 12:29

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

feat(annotation): panel-status core engine#248

feat(annotation): panel-status core engine#248
henrycgbaker merged 1 commit into
mainfrom
feat/annotation-retrieval-completeness/03-status-command

henrycgbaker commented Jun 1, 2026 •

edited

Loading

Uh oh!

ddimmery left a comment

Uh oh!

ddimmery Jul 4, 2026

Uh oh!

ddimmery Jul 6, 2026

Uh oh!

ddimmery Jul 6, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Uh oh!

Conversation

henrycgbaker commented Jun 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

ddimmery left a comment

Choose a reason for hiding this comment

Uh oh!

ddimmery Jul 4, 2026

Choose a reason for hiding this comment

Uh oh!

ddimmery Jul 6, 2026

Choose a reason for hiding this comment

Uh oh!

ddimmery Jul 6, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

henrycgbaker commented Jun 1, 2026 •

edited

Loading