← Back to V2 Schema Changes
V2 Schema Changes Record
Page Classifications · V2 Schema Changes · #27
id
27
ts
1778853426
kind
observation
before
blank (63K) + misc (17K) bucket combined ~80K
after
Estimated 60-70K (80%+) are misclassified image.photo
rationale
Confirmed across two batches: blank-b1 found 85/100 image.photo; blank-b2 found 75/95 image.photo. Combined batches: 160/195 = 82% are actually photos. Property-photo clusters: Little St. James construction (00542xxx, 0114-0117xxxx), NYC mansion interiors (00000xxx, 0055-0056xxxx). v1 pixel-density classifier broke entirely on photos with white borders. Recommend: per-doc CLIP visual-classifier pass on full 80K, expected ~65K → image.photo migration.
n docs touched
80000