Applied tags aren't scored live, so compute the grounding on demand: run the
tag's head over the image's max-over-bag (whole-image + concept crops), argmax
→ the region that best explains the tag on this image, mirroring what
score_image records for live suggestions.
- heads.py: extract _image_bag (now shared by score_image) + ground_applied_tag.
Returns (grounding, has_head): has_head False = no head to localize with →
no overlay; grounding None = the whole-image vector won → whole-image frame.
- tags.py: GET /api/images/<id>/tags/<id>/grounding → {grounding, has_head}.
- TagChip/TagPanel: applied chips inject fcSuggestionHover and fetch grounding
on hover (cached per image+tag, race-guarded), reusing Step 3's overlay in
both the modal and Explore. No new frontend overlay code.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM
score_image now keeps the ARGMAX beside the max-over-bag: which bag row won each
head. The region query also selects bbox/kind/detector_version, a parallel
bag_meta maps each row → its region (None for the whole-image vector), and every
hit gains grounding {bbox,kind,detector} (null when the global vector won). Threaded
through SuggestionService (new Suggestion.grounding field) → /api/.../suggestions
payload. This is the data the #1206 hover-overlay draws. CCIP-only hits ground null
for now (figure grounding = step 2). Tests: winning crop grounds the tag with its
bbox+kind; whole-image win → grounding None.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM
System tags are kind=general, so their suggestions previously landed in the
General group. Give them their own 'system' suggestion category so the operator
reviews them apart from content tags: _current_heads maps is_system heads to
category 'system' (still trained as general heads, still gated by the 0.65
floor). Frontend: CATEGORY_ORDER/LABELS gain 'system'; SuggestionsPanel renders
a 'System' group first (small, collapsible, open — false positives easy to spot
and reject); the typed-dropdown shows the shield icon for system entries. Safe:
system-tag suggestions always carry a canonical_tag_id, so the create-by-kind
path (which would send 'system' as a TagKind) is never hit.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM
System tags (wip/banner/editor) already get heads (kind=general) and aren't
filtered from suggestions, but they surfaced only at each head's precision-tuned
suggest_threshold — high enough to hide the borderline/false-positive guesses the
operator wants to SEE and REJECT (hard-negative mining: 'negatively reinforce
what isn't a system tag'). score_image now uses a flat _SYSTEM_TAG_SUGGEST_FLOOR
(0.65, operator-set) for system-tag heads instead of their auto threshold;
content-tag heads keep their own, and the typed-dropdown threshold_override still
overrides everything. _current_heads carries Tag.is_system into the head meta to
drive it.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM
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
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
SuggestionService.for_image now merges CCIP character matches with the SigLIP
head suggestions — they're complementary, not exclusive: CCIP is the identity-
specialized signal but needs a detected figure; the heads work whole-image but
conflate identity with style. Merged by tag: 'both' when they corroborate
(higher score wins), 'ccip' / 'head' otherwise. Cheap when no CCIP vectors exist
yet (match_image returns early without a figure vector), so it's a no-op until
the agent runs. Suggestion.source is now 'head' | 'ccip' | 'both'.
Test: a character with a CCIP reference figure surfaces (source='ccip') on a new
image whose figure matches.
NEXT: the agent container (real CCIP/detector models, hands-on) that produces the
vectors this consumes.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
The rail's Suggestions now come from the trained per-concept heads. SuggestionService.for_image scores the image's frozen SigLIP embedding against
every head (heads.score_image) and surfaces concepts above each head's own
suggest threshold; the typed-dropdown's min=0 "show everything" mode maps to a
flat floor so any head-scored concept can still be picked. Already-applied tags
drop; rejected tags stay flagged + reversible (unchanged).
REMOVED from the suggestion path (rule 22, no fallback): the Camie
ImagePrediction candidate/alias/merge pipeline and the per-tag centroid
augmentation, plus the now-dead SuggestionService internals (_load_predictions,
_threshold_for, _settings, self.aliases, self.centroids). Head suggestions are
always canonical tags, so raw_name/via_alias are null/false and the rail's
alias kebab is inert by data (its removal + the Camie ingest-tagger rip are the
flagged follow-up). for_selection (bulk consensus) now aggregates head
suggestions unchanged.
Tests rewritten to the head path: test_ml_suggestions (surfaces/applied/
rejected-reversible/override/no-embedding/no-heads), test_suggestions_bulk
(consensus), test_api_suggestions (get + dropped the Camie-alias roundtrip),
and test_ml_artist_retired (artist not head-eligible via _HEAD_KINDS).
DEPLOY NOTE: after this lands, the rail is empty until you run Train heads
(Settings → Tagging → Concept heads) — deploy, train, then the rail populates.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
test_rejected_tag_surfaced_flagged_then_reversible asserted "Rejectme" but an
existing tag keeps its stored name ("rejectme"), so the suggestion's
display_name is lowercase. Match by canonical_tag_id instead (casing-robust).
The feature was correct — only the assertion was wrong (run 1595 integration).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
A red-✗ dismissal no longer makes the suggestion vanish. The rejected tag
stays in the rail — dimmed, struck-through, with a "rejected" pill and a
one-click undo (↶) in place of the ✗ — so a misclick is recoverable and the
operator can see what they've said no to (operator-asked 2026-06-27).
Backend: SuggestionService.for_image now KEEPS rejected tags, flagged
rejected=True, sorted to the bottom of their category, instead of dropping
them. New AllowlistService.undismiss + POST /suggestions/undismiss clears the
TagSuggestionRejection. Rejected items are still excluded from bulk consensus
(for_selection) and the type-to-add dropdown, whose jobs are unchanged.
Frontend: store.dismiss flags in place (canonical tags) rather than dropping;
new store.undismiss reverts. SuggestionItem renders the rejected state and
swaps ✗→↶; ✓ still accepts (which clears the rejection server-side).
Tests: rejected-surfaced-flagged-then-reversible (service) + undismiss
endpoint idempotency (API).
Completes #1134's reversible-rejection half. Heads-as-suggestion-source is
the remaining piece.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
The headline bug: aliases created from the modal NEVER resolved. Create
sent the normalized display name ('Sword', 'Uchiha Sasuke') while
resolution keys on the raw booru model key ('sword', 'uchiha_sasuke',
case-sensitive) — so the mapping was stored under a key nothing looks up,
and the prediction kept reappearing unaliased. The raw key wasn't even in
the /suggestions response, so the modal couldn't send it.
- Suggestion now carries raw_name (the model key an alias must use) and
via_alias (surfaced via an operator alias); both serialized by the API.
- Modal alias-create sends raw_name, not display_name (the fix). Aliased
suggestions show an 'alias' badge and a 'Remove alias' action; 'Treat as
alias for…' is hidden for centroid hits (no model key) and already-aliased
rows.
- Tag-side management: TagCard ⋮ → 'Aliases…' opens a dialog listing the
model keys that fold into a tag, with remove (GET /api/tags/<id>/aliases +
AliasService.list_for_tag). Creation stays in the modal suggestion flow.
Tests: full API round-trip locking the raw-key contract (raw_name exposed →
alias authored with it → resolves + via_alias on a later image);
list_for_tag (service + API); via_alias/raw_name on the existing service
suggestion tests. No migration.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
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>
The typed dropdown sourced the threshold-filtered panel list (>= 0.70 general),
so low-confidence actions/features the model DID predict never appeared — forcing
hand-typed custom tags instead of accepting the model's canonical formatting.
Add a threshold override: SuggestionService.for_image(threshold_override=) and
GET /images/<id>/suggestions?min=<f> surface EVERY stored prediction (down to the
0.05 store floor), alias-resolved and normalized, still excluding applied/rejected
and unsurfaced categories. The suggestions store gains allByCategory + loadAll
(min=0); the dropdown searches that full set (cap 20), while the Suggestions panel
stays curated at the configured threshold. Accept/dismiss drop from both lists.
Operator-asked 2026-06-09. Test: a 0.30 general prediction is hidden by default
but surfaced with threshold_override=0.0; unsurfaced categories still excluded.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Camie's booru-style vocab strings (`uchiha_sasuke_(naruto)`,
`#unicus_(idolmaster)`, `1000-nen_ikiteru_(vocaloid)`, `:/`) were
surfacing raw in SuggestionsPanel — and worse, the SAME raw string was
written to tag.name on Accept, polluting the DB with `underscored_lowercase`
names that don't match the operator's "Title Case" tag convention.
Add backend/app/services/ml/tag_name.py with a single normalize()
applying nine rules (strip leading junk #/./+/;/~/_/ws, drop trailing
_(disambiguator) blocks iteratively, strip wrapping quotes, underscores
to spaces, space after colon, title-case each word's first char,
preserve hyphens/apostrophes/digits, drop entries with no letters).
Wire into SuggestionService.for_image:
- raw Camie key kept for alias_map lookup (alias rows are hand-curated
against raw keys; don't disturb)
- display_name = normalize(raw); None means drop the candidate
- existing-tag lookup widened to case-insensitive match against BOTH
raw and normalized forms so legacy underscore-named Tag rows accepted
before this change still surface as "existing" not "+ new"
The read path: load tagger_predictions, drop unsurfaced categories
(rating/meta/year), apply per-category thresholds, batch-resolve aliases,
skip applied + rejected, augment with centroid hits above the similarity
threshold, merge duplicate signals (take max score, mark source 'both'),
group by category, sort by score DESC. Tests marked integration.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
bvandeusen