e86b91dfe2adfbc118eaefedf922b7e80fa238e7
36 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 bvandeusen
|
3006e84cc0 |
feat(ml): soft auto-apply — retract auto-tags now below threshold (milestone 139)
Daily scheduled_retract_auto_tags re-scores standing auto-applied tags and drops the ones the model no longer supports: - retract_auto_applied_heads: per graduated head, re-score its source='head_auto' images (bounded — only the images already carrying the auto-tag, not the whole library) and remove ones now < auto_apply_threshold. - retract_auto_applied_ccip: per source='ccip_auto' character tag, max-cosine the image's figure vectors vs that character's prototypes; remove ones now below the ccip auto-apply threshold. Both SKIP operator-confirmed tags (TagPositiveConfirmation) and are SILENT — a low score isn't proof the tag was wrong, so no hard negative is recorded (that's reserved for an operator removal). No-op unless the relevant auto-apply switch is on. New daily beat. sklearn-free tests for both paths + the disabled no-op. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM |
||
|
bvandeusen
|
2cfbb284d5 |
feat(heads): incremental retraining — refit only changed tags (#1317 phase 2, m138)
train_all_heads is now incremental by default: a per-tag training-data fingerprint (positive + rejection count/latest-timestamp, stored on tag_head.train_fingerprint) means a manual Retrain refits ONLY the tags whose data changed — O(what you touched), not O(all heads). The nightly scheduled_train_heads passes full=True to reconcile sampled-negative + hygiene drift across every head. First incremental run after deploy still refits everyone (NULL fingerprints), stamping them, then it's incremental. The refit decision + fingerprint are split into sklearn-free helpers (_head_fingerprints, _heads_needing_retrain) so the incremental logic is unit-tested directly (train_head itself needs scikit-learn). Migration 0080. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM |
||
|
bvandeusen
|
a1ed53136e |
feat(ccip): refresh task + beat + retrain hook for character prototypes (#1317, m138 step 3)
- refresh_character_prototypes celery task wraps the incremental builder (sync ml worker); returns skipped / rebuilt=N removed=N. - Beat: every ~15 min (cheap global-gate no-op when idle) + a nightly full=True reconcile as belt-and-suspenders. - train_heads enqueues it on success, so the Retrain button AND the nightly head retrain refresh CCIP on the SAME trigger — unified lifecycle, as asked. The initial (cold) full build loads the whole reference set once in the background, never on a /suggestions request. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM |
||
|
bvandeusen
|
19b962f1a7 |
feat(b3): ml-worker becomes optional — embed-only role, decoupled GPU coordination, cpu-embed switch
The ml-worker's ONLY processing role is now the CPU whole-image embed fallback (tag_and_embed renamed embed_image — Camie tagging was retired #1189 and the name kept implying otherwise; videos were already handled agent-style: frame sampling + mean-pool). Detection/cropping/CCIP stay GPU-agent-only, and their completion is judged per-pipeline: ccip by gpu_job rows, siglip by concept regions at the current model version — never by image_record.siglip_embedding. A CPU embed therefore can NEVER close crop work for the agent (regression test pins this; only the whole-image 'embed' job, the same artifact, is satisfied). Making removal actually safe (operator will drop the container): - GPU-queue coordination (enqueue_gpu_backfill, recover_orphaned_gpu_jobs, reprocess_gpu_jobs) moved verbatim to tasks/gpu_queue.py on the maintenance quick lane — it lived on the 'ml' queue only by module colocation, which made the ml-worker a hard dependency of the whole agent pipeline. - New ml_settings.cpu_embed_enabled (migration 0074, default ON so agent-less installs keep working): OFF stops the four import hooks queueing embed work nothing will consume and no-ops the manual backfill; switch lives on the renamed 'CPU embedding backfill' card. - NB heads training / auto-apply still run on the ml image (sklearn) — a stack that removes the container gives those up too. Deploy note: in-flight messages under the old task names are dropped by the new workers; the 60s orphan sweep + hourly backfill re-fire under the new names immediately. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM |
||
|
bvandeusen
|
eaea4308fc |
chore: retire the tag-eval harness — it proved the heads system, job done (operator-approved)
The head-vs-centroid eval (#1130) existed to prove the 'frozen embedding + trained head' spine; the operator accepted the tagging system and dropped the harness. Removed per rule 22: TagEvalCard + store, /api/tag_eval blueprint, tag_eval_run ml task, recover-stalled-tag-eval-runs sweep + beat entry, TagEvalRun model + table (migration 0073), and its tests. The eval's data loaders + metric helpers were NOT eval-specific — the nightly heads trainer runs on them — so they moved verbatim to services/ml/training_data.py (heads.py import updated; behavior unchanged). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM |
||
|
bvandeusen
|
09e2772628 |
fix(gpu-jobs): end the error-tombstone loop — deliberate retry semantics + poison-job guards
The hourly ccip backfill's skip-list lacked 'error' (and the daily
siglip/embed variants re-gated failures on their missing results), so every
permanently-bad file got a fresh doomed job each run — ~24 duplicate error
rows/day per file, the perpetual 'unprocessable' flood. An errored job is now
a TOMBSTONE: no backfill re-enqueues it; retry is deliberate-only via
/retry_errors (an errored back-catalogue needs one button press after a
model swap).
One shared set of dedupe DELETEs (services/ml/gpu_jobs.error_dedupe_statements)
runs before every backfill and inside /retry_errors: error rows made moot by a
later pending/leased/done row go first, then older duplicates (newest reason
survives) — so the error count reads as distinct failing files and a retry
can't fan one file out into duplicate pending jobs. /retry_errors now returns
{requeued, pruned} and the toast shows both.
Poison-loop guards (release and lease-expiry burn no attempts, so a job that
stalls its transfer or crashes the agent every time cycled forever —
operator-observed jobs 99044/125288/131594/143131):
- agent: 3 in-session transient bounces (fetch or submit) → fail with the real
reason instead of another release; strikes never count while stopping, and
clear on submit success. Agent build 2026-07-02.3.
- server: the 60s orphan sweep (statements shared between the beat task and
GpuJobService so they can't drift) converts expired leases with >=5 lease
grants and pending jobs with >=10 to 'error', preserving the last stored
failure reason. Backstops old agent builds.
Tests: tombstone rule across all three backfill variants, moot-row pruning,
poison conversions, and the extended /retry_errors dedupe contract.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM
|
||
|
bvandeusen
|
80f8eb4756 |
feat(gpu): re-process trigger to apply new crop detectors to the existing library (#1202)
The siglip/ccip backfills skip images that already have current-version regions, so adding crop detectors only affected NEW images — the back-catalogue would never be re-cropped. Add a reprocess trigger that resets every done/error job of a task back to pending, so the agent re-runs the FULL pipeline (figure detection + CCIP + concept/panel crops) over the whole library under the current detectors. - reprocess_gpu_jobs(task='ccip') task + POST /api/gpu/reprocess. - gpu store reprocess() + GpuAgentCard "Re-process library (re-detect + re-crop)" button with a confirm (it's heavy). - Test: a done job resets to pending (attempts cleared). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa |
||
|
bvandeusen
|
485387ff0b |
refactor(ml): retire the Camie tagger + allowlist bulk-apply (#1189)
Heads + CCIP are the tag source and head auto-apply is the earned propagation.
The Camie tagger ran only to feed the allowlist bulk-apply (its ImagePrediction
rows had no other consumer), and the allowlist was a SECOND, un-earned auto-apply
path firing in parallel with heads on every accept — exactly the un-earned spray
the v2 pivot replaced. Retire both.
Behavior change: accepting a suggestion now applies the tag to THAT image only
(source='ml_accepted', a head-training positive) — it no longer allowlists +
fans the tag across the library via Camie. Propagation is heads' earned
auto-apply. (Loses instant cold-start propagation for booru-vocab tags; that was
un-earned and bypassed the precision gate.)
- tag_and_embed is now EMBED-ONLY (no Camie load/infer, no ImagePrediction
writes); backfill enqueues it for images with no embedding.
- Removed: services/ml/tagger.py, apply_allowlist_tags + helpers + daily beat +
every enqueue caller (accept/alias/merge/per-image), api/allowlist.py +
blueprint, ImagePrediction + TagAllowlist models/tables (migration 0067),
AllowlistTable.vue + allowlist store, the accept coverage-projection payload.
- AllowlistService gutted to accept/dismiss/undismiss/reject (the rejection store
the rail still needs); accept returns nothing, API returns {accepted, tag_id}.
- tag merge no longer repoints/triggers the allowlist; _keep_as_alias now keys on
ML-applied image_tag sources (incl. head_auto) instead of the allowlist.
- UI: MLBackfillCard relabelled to embedding-only; accept toast simplified;
MaintenancePanel drops the allowlist tile.
Left for a follow-up hygiene pass (now-inert, harmless): the dead settings
columns (tagger_store_floor, tagger_model_version, suggestion_threshold_*,
video_min_tag_frames), image_record.tagger_model_version, MLThresholdSliders
trim, and the Camie model download in download_models.py.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
|
||
|
bvandeusen
|
3d77a38a25 |
refactor(ml): remove the dead per-tag centroid subsystem (#1189)
The v2 pivot replaced per-tag SigLIP centroids with learned heads + CCIP. Centroids were still recomputed (on every tag merge + a daily beat) but NOTHING read them — suggestions come from heads+CCIP and apply_allowlist_tags applies via Camie predictions, not centroids. Pure dead wiring; remove it. Removed: CentroidService, recompute_centroid/recompute_centroids tasks, the daily beat, POST /api/ml/recompute-centroids, the recompute-on-merge trigger, the tag_reference_embedding table + model, the centroid_similarity_threshold + min_reference_images settings (migration 0066), the CentroidRecomputeCard + its store action + MaintenancePanel tile, and the centroid slider in MLThresholdSliders. _keep_as_alias drops its vestigial has-centroid branch (the allowlist branch already covers "could re-emit"); tag merge no longer clears a table that no longer exists. NOT touched (still live, parallel to heads): the Camie tagger, ImagePrediction, and the allowlist bulk-apply — accepting a suggestion still allowlists + applies it across the library. The tag-eval "centroid" baseline metric is unrelated (in-memory) and stays. (image_record.centroid_scores JSON column also remains — separate legacy field, its own micro-cleanup.) Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa |
||
|
bvandeusen
|
4daa3f2790 |
feat(ml): operator model swap — GPU re-embed + embedder as a setting (#1190)
Make the SigLIP embedder an operator choice (drop-in to SigLIP 2:
google/siglip2-so400m-patch16-512 is a verified 1152-d model at 512px → no
schema change, better small-cue fidelity). A swap = set model + re-embed +
retrain, all operator-driven; the GPU agent does the re-embed so it's fast.
- settings: embedder_model_name is now a setting (migration 0065) alongside the
existing embedder_model_version; both editable + validated (non-empty) in the
ml admin API. The server embedder loads by HF name (AutoImageProcessor/Model,
model-agnostic), preferring the pre-downloaded local dir for the default so
existing deploys don't re-download; rebuilds on a name change.
- agent: new 'embed' job = whole-image SigLIP embedding (mean-pool video frames)
under the lease-announced model → POST /jobs/submit_embedding writes
image_record.siglip_embedding + siglip_model_version. The lease now announces
the model FROM THE SETTING (not a constant).
- re-embed routing: enqueue_gpu_backfill('embed') selects unembedded + stale-
version images; 'siglip' now re-embeds concept crops whose version != current
(so a swap re-triggers crops, not just the never-embedded back-catalogue). The
CPU ml-worker backfill no longer re-embeds on a version mismatch (it can't
churn the library at 512px) — the GPU agent owns version re-embeds. Daily
'embed' + 'siglip' beats self-heal.
- scoring: score_image only bags embeddings in the CURRENT model's space (whole-
image gated by siglip_model_version, concept regions by embedding_version) so a
mid-swap stale vector isn't scored by new-space heads; legacy NULL = current.
- UI: GpuAgentCard "Embedding model (advanced)" — edit name/version, Save, and
"Re-embed library (GPU)" (queues embed + siglip); points at SigLIP 2.
Tests: lease announces model + submit_embedding round-trip; enqueue 'embed'
selects stale/unembedded; stale-version excluded from scoring; embedder model
settable + empty rejected; siglip gate updated to current-version concept.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
|
||
|
bvandeusen
|
c6f38b0dac |
feat(tagging): SigLIP concept crops + max-over-bag scoring (#114)
Lift recall on small/local concepts (glasses, cum, stomach-bulge, xray, lactation) that the whole-image SigLIP vector washes out: the GPU agent now embeds figure crops with SigLIP too, stored as kind='concept' regions, and the suggestion rail scores each image as a BAG (whole-image + every concept crop), taking each head's MAX over the bag. The whole-image vector is always in the bag, so this can never score lower than before. Model-agnostic by construction: the server ANNOUNCES the embedding model (HF name + version) in the lease, so the agent loads whatever the heads were trained in and stays in lock-step — a model swap is a server setting + a re-embed migration, never an agent change. - agent: model-agnostic CropEmbedder (torch/transformers get_image_features, fp16 on CUDA, inference-locked); worker branches on job.task — 'ccip' emits figure(CCIP)+concept(SigLIP) in one pass, 'siglip' emits concept-only so the back-catalogue backfill never churns figure/CCIP regions; torch cu124 + transformers in the image. - server: lease announces embed_model_name/embed_version; score_image is max-over-bag (version-filtered region embeddings); enqueue_gpu_backfill 'siglip' gates on a missing concept region (drains the back-catalogue, retries failures, no double-enqueue); daily siglip-backfill beat; UI button; /api/ccip/overview reports images_with_concept_siglip. - v1 scope: suggestion rail only — auto-apply stays whole-image (conservative; heads' thresholds were calibrated on whole-image). Bulk-apply bag = follow-up. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa |
||
|
bvandeusen
|
b91a230f12 |
feat(ccip): automation + reference quality — keep identity flowing hands-free (#114)
Works through the optional CCIP ideas + the "keep moving even if I forget" ask:
AUTOMATION (no button needed):
- Hourly beat auto-enqueues CCIP backfill — new images get embedded (and errored
ones retried) on their own; the queue never goes idle waiting for a click.
- CCIP auto-apply: a daily sweep tags confident matches (source='ccip_auto') so
identity tags keep flowing. ON by default (opt-out, like head auto-apply);
ml_settings.ccip_auto_apply_enabled + _threshold (0.92, above the suggest cut),
migration 0064. Vectorized (one matmul + reduceat per image), reversible, skips
already-applied/rejected. Switch + threshold in the GPU agent card; GET/PATCH
/api/ml/settings; auto_applied count in /api/ccip/overview.
REFERENCE QUALITY (the over-fire root cause):
- character_references now draws ONLY from single-character images — on a
multi-character image the tag is image-level, so every figure would otherwise
pollute each character's prototypes (a 2-char image tagged 'Velma' made
Daphne's figure a Velma reference). This is the contamination behind residual
over-firing.
- Cached on a cheap signature (char-tag count + ccip-region count/max-id) so the
reference load isn't redone on every modal open.
Tests: multi-character image not used as a reference; auto-apply tags a confident
match as ccip_auto.
NEXT (not done, confirmed): comic-panel cropping + SigLIP concept crops ("spot
interesting content").
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
|
||
|
bvandeusen
|
2cb0427868 |
feat(gpu): fast orphan recovery — graceful release + 60s sweep (#114)
So work an agent orphaned gets picked back up quickly, three layers: - GpuJobService.release(): a graceful agent stop hands its still-leased jobs back to pending instantly (POST /api/gpu/jobs/release), no waiting out the lease. - GpuJobService.recover_orphaned() + recover_orphaned_gpu_jobs Celery task on a 60s beat: resets expired leases (a hard-crashed agent) to pending and keeps the queue counts honest even when nothing is leasing. - Lease TTL 300→180s: still well above any single job (a capped-frame video embed is tens of seconds, and a live worker heartbeats), but a hard crash recovers faster once the sweep fires. Tests: release returns-to-pending (token-scoped), recover_orphaned resets only expired leases, release API round-trip. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa |
||
|
bvandeusen
|
558d965a1c |
fix(gpu): count backfill enqueues via RETURNING, not rowcount
result.rowcount is unreliable for INSERT…SELECT (returned -1), failing the idempotency assert. Use .returning(GpuJob.id) and count the rows. (run 1652) Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa |
||
|
bvandeusen
|
f247f9247c |
style(gpu): ruff — split as-import, dict(rows) over comprehension
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa |
||
|
bvandeusen
|
6cabef07a4 |
feat(gpu): HTTP job API + token auth + backfill — the agent's server side (#114 slice 3b)
The thin HTTP surface over the queue so the desktop agent stays HTTP-only: - Agent endpoints (Authorization: Bearer <token>): POST /api/gpu/jobs/lease (returns jobs + image_url + mime + video frame cadence), /submit (stores regions via RegionService + closes the job; 409 on a stale lease), /heartbeat, /fail. Token validated against AppSetting (mirrors the extension-key pattern, constant-time compare). - Admin (browser): GET/POST /api/gpu/token[/rotate] (generate + show the agent token), GET /api/gpu/status (queue counts), POST /api/gpu/backfill → dispatches enqueue_gpu_backfill. - enqueue_gpu_backfill(task): one INSERT…SELECT enqueues a job per image lacking one for the task (scales to the full library; idempotent). Agent flow: lease over HTTP → fetch pixels via the normal FC image URL → compute on the GPU → submit. Redis/Postgres never exposed. Tests: bearer required (+ wrong-token 401), lease→submit round-trip (region+CCIP vector stored, job done via /status), stale-lease 409, backfill enqueue + idempotency. NEXT: the agent container + control UI, then the CCIP detector/embedder + matcher. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa |
||
|
bvandeusen
|
74fef908d2 |
feat(heads): earned auto-apply — sweep mechanism, off by default (#114 auto-apply A)
Graduated heads can now apply their tag without a human — gated so it's safe:
- FIRING GATE: a head fires only when the master switch (head_auto_apply_enabled,
default OFF) is on AND it has >= head_auto_apply_min_positives (default 30)
clean labels. A precise-looking but under-supported low-N head can't spray tags.
- auto_apply_sweep (heads.py): streams every embedded image in chunks, scores
against the eligible heads (numpy, no sklearn), applies each head's tag where
score >= its auto_apply_threshold and the tag isn't already applied/rejected,
with source='head_auto' (distinguishable + reversible). dry_run counts only.
- HeadAutoApplyRun (migration 0059) tracks each sweep / preview; apply_head_tags
task (ml queue) + scheduled_apply_head_tags daily beat (no-op unless enabled)
+ recovery sweep + retention(20).
- API: POST /api/heads/auto-apply {dry_run} (202 / 409 running / 400 disabled),
GET /api/heads/auto-apply (recent runs + per-concept report). Settings
head_auto_apply_enabled + min_positives via /api/ml/settings.
Tests: sweep applies above threshold, dry-run writes nothing, skips under-
supported + ungraduated heads; API disabled/dry-run/conflict guards.
NEXT (slice 2): the observability the operator asked for — per-concept misfire
(auto-applied-then-removed) + under-fire tracking, time-series snapshots, and a
reporting API to tune. Slice 3: the UI (enable, preview, trends).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
|
||
|
bvandeusen
|
77baee49fd |
feat(heads): nightly auto-retrain + inline Retrain button in Explore
Two cadences for keeping heads in sync with your tagging: - PASSIVE: a nightly `scheduled_train_heads` beat (skips if a run is already in flight; creates+commits the run row before dispatching train_heads so the ml worker always finds it). Folds the day's accepts/rejects + newly-eligible concepts into the heads without anyone clicking. - ACTIVE: a "Retrain heads" button in the Explore trail bar — bank the +/- feedback you just gave while walking content, without a trip to Settings. Shared logic in a new useHeadTraining composable (trigger + poll + start/finish toasts), used by the Explore button; reflects an already-running run (incl. the nightly one) on mount. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa |
||
|
bvandeusen
|
22c3b54746 |
feat(heads): production per-concept heads — train + score backend (#114 A)
The eval (#1130) proved the frozen-embedding + trained-head spine; this lands its production form (the first of three slices that make heads the suggestion source, replacing Camie + centroid). - tag_head: one logistic-regression head per general/character concept with enough labelled positives. Weights (pgvector), honest CV-derived suggest threshold + earned-auto-apply point, and per-concept quality metrics. - head_training_run: persisted batch lifecycle (mirrors tag_eval_run) so the admin card shows live + historical status across navigation. - services/ml/heads.py: TRAIN (sync, ml worker, reuses tag_eval's proven data loaders + metric math so production heads match measured eval numbers) and SCORE (async, API worker — numpy via pgvector, no scikit-learn): score one image's embedding against all heads → the rail's suggestions, cached on (count, max trained_at) so a retrain invalidates without per-request loads. - tasks.ml.train_heads (ml queue, commits per head so a kill leaves progress) + recover_stalled_head_training_runs sweep + retention(20) + 5-min beat (rule 89). - api/heads.py: POST /api/heads/train (one run at a time, 409 guard) + GET /api/heads (count, graduated, last-trained, running, per-concept table, recent runs). - ml_settings: head_min_positives + head_auto_apply_precision, tunable via /api/ml/settings. Scoring isn't wired into the rail yet (slice C) and the admin UI is slice B — this slice makes training + scoring exist and CI-verifiable. 'precision' column stored as precision_cv (SQL reserved word). Migration 0058. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa |
||
|
bvandeusen
|
6e3c5f697f |
feat(ml): tag-eval backend — head-vs-centroid learning-curve eval (persisted)
Slice 1 of milestone #114 (tagging v2). Proves the frozen-embedding + trained- head spine on the operator's own data, reusing the SigLIP embeddings already stored on image_record — no re-embedding, no GPU. Per concept: train a logistic-regression HEAD (positives + negatives = explicit rejections + sampled unlabeled) vs the old single-CENTROID baseline; report cross-validated precision/recall/AP for both, a LEARNING CURVE (AP/F1 as tagged positives grow 10→30→100→300), and example image ids (head-would-suggest / head-doubts-positive) to eyeball. Persisted so the report SURVIVES navigation (operator-flagged): the run + full report live in a new tag_eval_run row (mirrors library_audit_run); the admin card will rehydrate from GET on mount, not transient state. - models.TagEvalRun + migration 0056; runs on the ml queue (only worker with numpy/sklearn) — numpy/sklearn lazy-imported so the API can still enqueue. - services/ml/tag_eval (compute + start helper, one-running guard), tasks.ml .tag_eval_run, api/tag-eval (POST create, GET history light / detail w/ report). - recover_stalled_tag_eval_runs sweep + retention (keep last 20) + 5-min beat (rule 89). scikit-learn added to requirements-ml. - tests: param normalization + the rehydrate read-path + create/conflict. Frontend admin card (trigger + render persisted report) follows next. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
||
|
bvandeusen
|
51201b459e |
fix(ml): per-task async engine for recompute_centroid (#881)
recompute_centroid + recompute_centroids were the only tasks still using the process-wide singleton extensions.get_session() under asyncio.run(). The async engine's asyncpg pool is bound to the loop it was created on; each Celery task runs a fresh asyncio.run() loop, so after the first invocation the cached engine handed loop-A connections to loop B and raised "Future attached to a different loop" — every recompute after the first in a worker process failed (~35ms, fails on first DB await). Convert both to the established per-task async_session_factory() pattern (NullPool engine created + disposed inside the task's own loop), matching scan/download/admin tasks. No get_session usages remain in tasks/. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> |
||
|
bvandeusen
|
369e3de684 |
feat(ml): cadence-based video frame sampling + min-frame tag aggregation (#747)
Video tag noise root cause: frames were a FIXED count (6) max-pooled — a tag firing on one frame survived at peak confidence, and a fixed count under-samples long multi-scene videos so real scene-local tags looked like noise. Redesign (operator-steered): - Sample at a fixed CADENCE — one frame every `video_frame_interval_seconds` (default 4) across the 5–95% window — so a tag's frame-presence reflects real screen time independent of video length. Capped at `video_max_frames` (default 64): a long video stretches the spacing instead of exploding into hundreds of inferences, bounding per-video cost on the single ml-worker (per-frame ffmpeg timeout also cut 60s→30s). - Aggregate with `_aggregate_video_predictions`: keep a tag only if it appears in >= `video_min_tag_frames` sampled frames (≈ that many × interval seconds on screen — duration-independent noise rejection), with confidence = MEAN over the frames it appears in (not max). Clamps the threshold to the sample count so a 1–2-frame short video still tags. - All three knobs are DB-backed ml_settings (migration 0053), patchable via /api/ml/settings + sliders in the ML settings card — replaces the VIDEO_ML_FRAMES env var (product-not-project). Tests: aggregation drops one-frame noise + means corroborated tags + clamps on short videos; settings round-trip + min>max validation. Replaced the _maxpool_predictions unit test. NOTE: this is the QUALITY half of #747. The perf half — the ml-worker runs CPU-only — is GPU enablement, tracked separately in #872. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> |
||
|
bvandeusen
|
3610ba495f |
feat(ml): drop image_record.tagger_predictions — image_prediction is sole store (#768 step 3)
Read cutover verified in prod (suggestions + allowlist read image_prediction; backfill complete at 908k rows / 51k images). Removes the old JSON column and everything that fed it: - ImageRecord.tagger_predictions column removed; migration 0046 DROPs it. tagger_model_version kept as the "tagged / current?" signal the backfill sweep reads (needs-tagging check switched to tagger_model_version IS NULL). - tag_and_embed no longer dual-writes the JSON — image_prediction is the only write path. - importer re-import reset drops the JSON line (image_prediction rows are already deleted on re-import). - Retired the one-time #768 backfill task + the #764 prune task, their admin endpoints, and their Maintenance cards (Backfill/PrunePredictionsCard). - Tests seed/assert via image_prediction; stale column refs removed. Disk reclaim is NOT automatic: DROP COLUMN is a catalog change. Run `VACUUM FULL image_record` off-hours afterward to return the ~100 GB to the OS so DB backups go small (#739). image_prediction (~90 MB) stays in pg_dump — it's the source of truth now. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
||
|
bvandeusen
|
22cdf0f334 |
feat(ml): read suggestions + allowlist from image_prediction (#768 step 2)
Switch every prediction READER off the JSON column onto the normalized
image_prediction table. Parity by construction: each reader loads the same
{raw_name: {category, confidence}} dict it consumed before (via small
_load_predictions helpers), so all downstream threshold/alias/merge/consensus
logic is byte-identical — only the data source changed.
- suggestions.SuggestionService.for_image (and for_selection via it)
- ml.apply_allowlist_tags (iterates images that have prediction rows)
- importer re-import reset deletes the image's prediction rows
The tagger_predictions JSON column is still dual-written (step 1) so it stays
valid during transition; the backfill task's NULL check still works. Removing
the JSON write + DROP column + retiring the #764 prune is the cleanup
follow-up (needs a quiesced-worker window for the DROP lock).
Tests: shared tests/_prediction_helpers.seed_predictions seeds the table;
read-path tests (suggestions, bulk consensus, allowlist apply, API) seed there
instead of ImageRecord.tagger_predictions.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
||
|
bvandeusen
|
79089b50b0 |
feat(ml): image_prediction table + backfill + dual-write (#768 step 1)
Normalize tagger predictions out of the image_record.tagger_predictions JSON blob into a queryable per-prediction table. Step 1 of the cutover (expand): additive + low-risk — reads still use the JSON, this just adds the table and keeps it populated. - ImagePrediction(image_record_id, raw_name, category, score) — stores the RAW tagger vocab name (not tag_id) so read-time alias→canonical resolution is unchanged. Indexed for per-image reads + by (raw_name, score). - Migration 0045: create table + set-based backfill from the JSON via json_each (fast post-#764-prune). The old column stays (vestigial) and is dropped in a later follow-up — DROP needs an ACCESS EXCLUSIVE lock on the hot image_record table, so it waits for a quiesced-worker window. - tag_and_embed dual-writes the rows (delete-then-insert, idempotent); tagger_store_floor already applied in infer(). Next: switch suggestion + allowlist reads to the table, then drop the JSON write. Plan-task #768. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
||
|
bvandeusen
|
3f92669f12 |
feat(ml): DB-backed tagger_store_floor (default 0.70), the ingest confidence floor
Promotes the prediction store-floor from the TAGGER_STORE_FLOOR env (default 0.05) to a DB-backed, Settings-UI-tunable ml_settings column (default 0.70). Storing every tag down to 0.05 from a ~10k-tag tagger is what grew image_record's TOAST to ~100 GB; the suggestion path already filters at 0.70 and the centroid/learned path covers lower-confidence preferred tags, so the sub-0.70 tail is redundant. Foundation for plan-task #764 (backfill + reclaim land next; this only changes the write gate for NEW imports). - ml_settings.tagger_store_floor (migration 0044, default 0.70) - tagger.Tagger.infer(store_floor=...); ml task passes settings.tagger_store_floor - ML admin GET/PATCH expose it; PATCH rejects a category suggestion threshold below the floor (nothing below the floor is stored, so the gap surfaces nothing) — server backstop for the UI slider clamp - Settings → ML: store-floor slider + caption; category sliders min-bound to it Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
||
|
bvandeusen
|
f2fbe2ae6e |
tweak(ml): default video frame samples 10 to 6
Operator: 10-frame max-pooled tagging on video produces a lot of noisy tags, and the sampling burns time/GPU. Drop the VIDEO_ML_FRAMES default to 6 (still env- overridable). Fewer frames = less per-frame noise into the max-pool and a smaller frame-sampling budget. Quality/perf of the whole video path is being reviewed separately. |
||
|
bvandeusen
|
b1778ca9f2 |
obs(ml): tag_and_embed logs file + phase + timing; failures name them
The task logged nothing and SoftTimeLimitExceeded stringifies to empty, so a timeout surfaced as a bare 'SoftTimeLimitExceeded()' with no clue which file or why (operator-flagged 2026-06-08). - Log start (id/path/mime/bytes/video?), per-phase timing (load_models, video probe/sample/infer, tag, embed, persist), and a success summary. - Track a + file ; on SoftTimeLimitExceeded log it and re-raise SoftTimeLimitExceeded WITH that context (keeps the 'timeout' task_run status but gives the activity a real error_message: which file, which phase, elapsed). - On other exceptions, log context then re-raise the ORIGINAL (preserves autoretry for OSError/DBAPIError/OperationalError). Now a stuck run names the culprit — most likely a slow video (frame sampling is up to 10x60s ffmpeg) or a huge image; the phase log will say which. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
||
|
bvandeusen
|
e30f50e6fe |
fix(audit-g3): lifecycle batch — recovery sweeps, retention, timeouts
Plugs the FC long-running-entity discipline gaps the 2026-06-02 audit flagged: every entity that can get stuck now has recovery + retention + timeout, and the long-runners no longer collide with the FC-3i sweep. Recovery sweeps (every 5 min): - recover_stalled_backup_runs — flips BackupRun stuck in running/restoring past 7h (covers the 6.5h images-backup hard limit) to error. prune_backups docstring corrected — the FC-3i TaskRun sweep never touched BackupRun rows. - recover_stalled_library_audit_runs — flips LibraryAuditRun stuck past 135 min (10-min buffer above scan_library_for_rule's 2h5m hard limit) to error. Previously a SIGKILL'd row blocked all future audits until manual DB surgery. - recover_stalled_import_batches — finalizes ImportBatch rows stuck running >2h whose child tasks are all terminal (orphan case where the orchestrator crashed before the closing UPDATE). Uses the same EXISTS predicate /api/system/stats already had. Retention (daily): - prune_library_audit_runs — 30-day window. Audit rows carry matched_ids JSONB blobs that can hold tens of thousands of ids. - prune_import_batches — 30-day window. Cascades to ImportTask via the model relationship. time_limits on five long-runners that previously had none (the audit's headline finding — every one of these collided with the recover_stalled_task_runs 5-min default and could be marked 'error' mid-flight): - scan_directory: 60m soft / 70m hard - verify_integrity: 60m / 70m - backfill_phash: 30m / 35m - apply_allowlist_tags: 30m / 35m - recompute_centroids: 30m / 35m QUEUE_STUCK_THRESHOLD_MINUTES now covers maintenance (75) and scan (75) — above the longest task on each — with per-task overrides for the outliers (backup_images_task 420, restore_images_task 420, scan_library_for_rule 130). start_audit_run guard is now age-aware: a 'running' row older than the audit hard limit doesn't block a new run (the sweep will catch it within 5 min). Previously a SIGKILL'd row blocked forever. /api/import/status now uses the same EXISTS predicate /api/system/stats does, so the two endpoints no longer disagree on the active-batch question. DownloadEvent.started_at resets on pending→running so a freshly- promoted event from a busy queue isn't measured against its original enqueue time (was racing recover_stalled_download_events on heavy-queue days). |
||
|
bvandeusen
|
e3cdd0f92b |
feat(import-resilience L3): subprocess-isolated probes for video + archive
Layer 3 — prevent the hard worker crash rather than just recovering from it. The realistic process-crash vectors (operator's observed slow/heavy tasks) are video decode and archive extraction; images decode in-process and Pillow raises-and-skips cleanly, and a subprocess per image would wreck deep-scan throughput, so images are intentionally not probed. New backend/app/utils/safe_probe.py (leaf module, lazy heavy imports so the spawned child stays light): - probe_video(path): validates the container + first video stream via ffprobe (a separate binary — a decoder crash kills only ffprobe, not the worker). Returns width/height, which the importer didn't capture for videos before. crashed=True only on ffprobe timeout. - probe_archive(path): an uncompressed-size bomb guard (MAX_ARCHIVE_UNCOMPRESSED_BYTES = 4 GiB) plus the format integrity test (zipfile.testzip / rarfile.testrar / py7zr.test) run in a spawned child process. A decompression-bomb OOM or native-lib segfault on a malformed archive shows up as a non-zero child exit code → crashed=True, never a dead worker. ProbeResult.crashed distinguishes a HARD failure (subprocess killed / timed out — the poison-pill signature → caller returns terminal 'failed') from a CLEAN rejection (corrupt-but-handled, bomb cap, integrity mismatch → caller's choice of skipped/attached). Wired: - importer._import_media video branch: probe_video before the pipeline; crash → failed, clean reject → invalid_image skip, ok → capture dims. - importer._import_archive: probe_archive before extract_archive; crash → failed, clean reject → still preserve the archive as a PostAttachment (matches extract_archive's fail-soft contract). - ml.tag_and_embed video branch: probe_video before sampling 10 frames, so a corrupt video is rejected (status='bad_video') instead of crashing the ml-worker on frame decode. Tests (test_safe_probe.py): valid/corrupt zip via probe_archive, direct _inspect_archive size+integrity, in-process _archive_probe_target bomb guard (monkeypatch can't reach a spawned child, so the target is called directly), and a non-video → ok=False that's robust to ffprobe presence in CI. |
||
|
bvandeusen
|
407de18ff6 |
fix(ml): video branch needs longer time limits; recovery sweep is now per-queue
Operator-flagged 2026-05-28: tag_and_embed on image 6288 (an mp4) was
marked failed by recover_stalled_task_runs at the 5-min sweep tick
while still legitimately running. The error_type='RecoverySweep' /
"no completion signal received within 5 min" message was misleading
— the worker was busy, not stuck.
Root cause is two interacting limits, both undersized for video work:
tag_and_embed: soft_time_limit=300, time_limit=420
(sized for the image branch, ≈2 GPU ops)
recovery sweep: STUCK_THRESHOLD_MINUTES = 5 across all queues
The video branch samples 10 frames via ffmpeg, then runs tagger +
embedder on EACH frame — ~20 GPU ops vs 2 for an image. A loaded
ml-worker can take 5-10 min on a long video, which trips both
limits well before the task naturally finishes.
**Two-part fix**
1. `tag_and_embed` time limits bumped to soft=900 (15 min) / time=1200
(20 min). Sized for the video path's worst case; image runs return
in seconds and don't care.
2. New `QUEUE_STUCK_THRESHOLD_MINUTES` override dict in maintenance.py.
Queues with legitimately-long-running tasks (currently just `ml` at
25 min — 5-min buffer past the new hard kill) get their own
threshold; queues not in the dict use the default 5 min. The sweep
now issues one UPDATE per distinct threshold value, with
`queue.notin_(override_queues)` on the default pass so each row is
touched at most once.
Tests:
- _make_task_run helper accepts `queue=` (defaults to "default") so
existing tests use the default-threshold path.
- New test `test_recover_stalled_task_runs_ml_queue_uses_longer_threshold`
pins both directions: a 10-min-old ml row survives (fresh by 25-min
override), a 30-min-old ml row gets flagged.
After deploy, operator's mp4 ML jobs run to completion without
spurious RecoverySweep failures.
|
||
|
bvandeusen
|
be0f472894 |
fix(workers): worker-level recovery — autoretry on transient errors + tightened sweep (5min) + import_media_file body-wrap so no path leaves rows stuck in 'processing'
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
bvandeusen
|
7d8b9c3d90 |
fix(tasks): share one sync engine per worker process to stop connection-pool leak
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
bvandeusen
|
f7f75fcac6 | refactor(fc2c-i): sweep blueprints + ml tasks onto shared get_session (covers tasks/ml.py, a survey gap) | ||
|
bvandeusen
|
3e6cc8fffa |
feat(fc2b): add allowlist-apply + centroid recompute tasks + beat
apply_allowlist_tags: 4 modes (tag-only / image-only / both / full sweep), matches a tag to a prediction either by direct name or via alias (name, category) resolution, gates on per-tag min_confidence, skips applied/rejected, applies source='ml_auto'. recompute_centroid / recompute_centroids: async-bridged calls into CentroidService, delta-gated. Beat: daily backfill, daily centroid recompute, daily allowlist sweep. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|
bvandeusen
|
ac7e0d13bc |
feat(fc2b): add tag_and_embed + backfill Celery tasks
tag_and_embed: Camie + SigLIP on one image (video → 10-frame sample, max-pool tags, mean-pool embeddings), stores predictions/embedding with model versions, then enqueues per-image allowlist apply. backfill: keyset-paginated discovery of images missing predictions/embeddings for the current model versions (restart-safe). apply_allowlist_tags stub included so .delay() resolves between commits (filled in Task 9). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |