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

docs(decision): re-measure E-3 and fix the labels README's dead reference

Browse source 
E-3 with the calibrated lines: 66% at 95% precision with the text lane, 55% at
81% without, 2 false accepts against 8. The lane adds 11 pp of coverage while
raising precision 14 pp, so the answer stands and the margin is wider than the
audit's.

The labels README pointed at evidence/image6_v2.json, which was a byte-identical
duplicate of eval/labels/image6.json and is now removed; the README says so
instead of describing a copy that no longer exists.
This commit is contained in:
kiyo9w 2026-09-29 04:31:59 +09:00
parent 4d1c17f15d
commit 684934e042
2 changed files with 10 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

5
recognization/docs/research/2026-09-28-recognition-architecture/DECISION.md
View file

@ -815,6 +815,11 @@ against 20). Live: 62% at 93% against 56% at 84% (3 against 7). The lane adds
16 pp and 6 pp of coverage while *raising* precision, against E-3's +5 pp bar. 16 pp and 6 pp of coverage while *raising* precision, against E-3's +5 pp bar.
It is not a veto-only lane. It is not a veto-only lane.
Re-measured 2026-09-29 with the calibrated accept lines: 66% at 95% precision
with text, 55% at 81% without (2 false accepts against 8). The lane still adds
11 pp of coverage while raising precision by 14 pp, so the answer is unchanged
and the margin is wider than the audit's.
**E-2 answered.** `LocalLane` is RootSIFT-only; ORB and BRISK survive only in **E-2 answered.** `LocalLane` is RootSIFT-only; ORB and BRISK survive only in
the `matcher` baseline lane, so there is nothing to drop from the shipped path. the `matcher` baseline lane, so there is nothing to drop from the shipped path.

9
recognization/eval/labels/README.md
View file

@ -30,9 +30,10 @@ complete; every scorer drops them before computing a rate. The evaluators read
in-catalog across 12 SKUs, 45 `unknown`. 26 labels changed from v1, and three in-catalog across 12 SKUs, 45 `unknown`. 26 labels changed from v1, and three
more changed on 2026-09-29: facings #46-48 were `unknown` with the note more changed on 2026-09-29: facings #46-48 were `unknown` with the note
"possibly what stock.csv calls l12", and the l12 reconciliation settled it, so "possibly what stock.csv calls l12", and the l12 reconciliation settled it, so
they now carry `sku_l12`. It is a copy of they now carry `sku_l12`. This file is the single copy: the evidence directory
`docs/research/2026-09-28-recognition-architecture/evidence/image6_v2.json`; used to hold a duplicate (`image6_v2.json`) and it was removed on 2026-09-29
edit the evidence copy and re-copy, so the two cannot drift. after `cmp` showed the two byte-identical, because a copy that nothing reads is
a copy that drifts.
## Schema check ## Schema check
@ -50,4 +51,4 @@ are absent rather than guessed.
Tier-2 labels must add both. `confidence` is the labeler's own certainty for Tier-2 labels must add both. `confidence` is the labeler's own certainty for
that row; `labeler` names the person or pass that produced it, because Tier 2 is that row; `labeler` names the person or pass that produced it, because Tier 2 is
double-labelled and a disagreement has to be traceable to a pass. Tier 1 double-labelled and a disagreement has to be traceable to a pass. Tier 1
(`image6_v2.json`) is dev-only and its numbers never ship, so it stays as it is. (`image6.json`) is dev-only and its numbers never ship, so it stays as it is.