tngo/product-recognition-demo
1
0
Fork 0
forked from vnguyen/product-recognition-demo
Code Pull requests Activity

fix: stop re-gating gallery scores in the live path

Browse source 
`score_crop` returns only scores the gallery already accepted: the raw 0.55
line, or the fine-tuned head's own 0.40. The live path then ranked that
aggregate with `decide_identity(scale="cosine")`, whose defaults rewrite to
COSINE_ACCEPT 0.60 / COSINE_AMBIGUOUS 0.50 - a second, higher line. Every head
match between 0.40 and 0.60 was therefore refused or downgraded to ambiguous,
which is the live regression: the head's own training views score 0.57-0.77.

`engine.inspect` and `_classify_with_gallery` already avoid this by ranking
through `decision_for_match`, which applies only the published margin. The live
path now does the same: on the cosine scale it passes explicit zero accept
lines, so the margin gate is the only extra gate.

Regression tests: a 0.45 head-scale match is labelled (fails on the previous
code), and a 0.45/0.44 pair inside the margin stays unnamed.

Verified: `python -m unittest discover -s code -p 'test_*.py'` - 257 tests, OK,
2 skipped.
This commit is contained in:
kiyo9w 2026-09-28 17:34:47 +09:00
parent 1bb1b7158b
commit 84ca113b90
2 changed files with 57 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

14
recognization/code/live.py
View file

@ -1146,9 +1146,17 @@ class LiveSession:
from identity import aggregate_votes, decide_identity
mean_scores = aggregate_votes(track.observations)
decision = decide_identity(
mean_scores, method=method, scale=getattr(self.engine, "identity_scale", "inlier")
)
if getattr(self.engine, "identity_scale", "inlier") == "cosine":
# score_crop only returns a score the gallery already accepted, so
# the aggregate must not be re-gated at COSINE_ACCEPT: that line is
# above the head's own 0.40 and refused matches the gallery had
# accepted. Only the margin gate is left to apply.
decision = decide_identity(
mean_scores, method=method, scale="cosine",
min_score=0.0, min_confidence=0.0,
)
else:
decision = decide_identity(mean_scores, method=method)
track.candidates = [
{"sku_id": row.get("sku_id"), "score": row.get("score")}
for row in (decision.get("ranked_candidates") or [])[:3]