chore(eval): reconcile the distractor sets and refresh the sweep numbers

The Wikimedia collector truncated the whole basename at 60 characters, so 13
files lost their extension and cv2 could not read them by name - the distractor
sets were quietly scoring 13 images fewer than they held. It now truncates the
stem and keeps the extension, and the files on disk are renamed to match.

The search terms also pull in .djvu volumes whose page 1 Commons serves as a
JPEG thumbnail. Twelve book scans were in the sets; they contributed 0 false
accepts, and they are not shelves, so the collector filters them and the
manifests no longer list them.

With all 2650 crops readable the sweep is re-measured end to end: 3.09% at the
old lines, 0.30% at the shipped 4.0/4, image6 44 shown / 42 correct / 95.5%.
The accept sweep table and the verdict's far-unknown sentence now carry those
numbers instead of the pre-calibration ones.

Also: the fetched images are gitignored (manifests and results are not), the
l10/l11 shelf-tag crops that back the catalog corrections are committed, and a
stale pre-gate eval.json and a duplicate of eval/labels/image6.json are gone.
This commit is contained in:
kiyo9w 2026-09-29 04:17:54 +09:00
parent 4bdad66bbe
commit 5eb4bdf6c8
19 changed files with 646 additions and 253 deletions

View file

@ -25,3 +25,14 @@ stock-*/identity_head.pt
.gallery_cache_*.npz
.demo-token
config/local.env
# Fetched images. Every set keeps its manifest.json (source URL, licence,
# artist) and its results in git; the pixels are re-fetchable from those URLs
# and are 59 MB for the Wikimedia shelves alone.
data/external-shelves/*/*.jpg
data/external-shelves/*/*.jpeg
data/external-shelves/*/*.JPG
data/external-shelves/*/*.png
data/packshots/research/*/images/
data/packshots/research/*/pages/
data/packshots/research/*/*.jpg