product-recognition-demo/recognization/data/external-eval/other-mayo/eval_result.json
kiyo9w 11826cc43c feat(eval): report the external benchmark per SKU
A set total hides which products the lane can and cannot name. Every SKU in
these sets has images from two independent source domains, so a SKU that scores
the same on both is not one lucky photograph.

The breakdown is what makes the two interesting results visible: l01, l02, l03,
l07 and l09 are 2/2 correct on both their domains, l07 with no enrolled image
at all, while l11 abstains on both and l04 abstains on all four of its images.
2026-09-29 03:37:04 +09:00

101 lines
No EOL
1.8 KiB
JSON

{
"dataset": "other-mayo",
"catalog": "stock-lopia-clean",
"note": "External proxy images, not Tier 2. Scored through the shipped LocalLaneAdapter + DecisionPolicy.",
"rows": [
{
"file": "sku_l04_1.jpg",
"truth": "sku_l04",
"best": null,
"confidence": 0.0,
"status": "rejected",
"source": null,
"verdict": "abstain"
},
{
"file": "sku_l04_2.jpg",
"truth": "sku_l04",
"best": null,
"confidence": 0.0,
"status": "unknown",
"source": null,
"verdict": "abstain"
},
{
"file": "sku_l04_3.jpg",
"truth": "sku_l04",
"best": null,
"confidence": 0.3294,
"status": "ambiguous",
"source": null,
"verdict": "abstain"
},
{
"file": "sku_l04_4.jpg",
"truth": "sku_l04",
"best": null,
"confidence": -1.6712,
"status": "unknown",
"source": null,
"verdict": "abstain"
},
{
"file": "sku_l07_1.jpg",
"truth": "sku_l07",
"best": "sku_l07",
"confidence": 0.0,
"status": "success",
"source": "text",
"verdict": "correct"
},
{
"file": "sku_l07_2.jpg",
"truth": "sku_l07",
"best": "sku_l07",
"confidence": -0.5503,
"status": "success",
"source": "text",
"verdict": "correct"
},
{
"file": "sku_l09_1.jpg",
"truth": "sku_l09",
"best": "sku_l09",
"confidence": 27.9546,
"status": "success",
"source": "local",
"verdict": "correct"
},
{
"file": "sku_l09_2.jpg",
"truth": "sku_l09",
"best": "sku_l09",
"confidence": 0.975,
"status": "success",
"source": "text",
"verdict": "correct"
}
],
"scored": 8,
"leaked": 0,
"by_sku": {
"sku_l04": {
"correct": 0,
"abstain": 4,
"wrong": 0
},
"sku_l07": {
"correct": 2,
"abstain": 0,
"wrong": 0
},
"sku_l09": {
"correct": 2,
"abstain": 0,
"wrong": 0
}
},
"correct": 4,
"abstain": 4,
"wrong": 0
}