Doug Lito: raw computed results
The raw ranking from the production similarity calculation (face-embedding centroid cosine → gender-cohort z-normalization). The web-opinion crown, curation, and fusion re-rank are removed.
- 1
96.6%z 6.08 - 2
83.2%z 4.28 - 3
81.6%z 4.15 - 4
81.1%z 4.11 - 5
79.3%z 4.01 - 6
75.3%z 3.97 - 7
73.9%z 3.95 - 8
69.3%z 3.90 - 9
67.5%z 3.88 - 10
67.3%z 3.88 - 11
66.7%z 3.87 - 12
63.2%z 3.83 - 13
59.4%z 3.79 - 14
59.1%z 3.79 - 15
55.8%z 3.75 - 16
52.3%z 3.71 - 17
50.7%z 3.70 - 18
49.8%z 3.69 - 19
49.1%z 3.68 - 20
48.9%z 3.68 - 21
47.6%z 3.64 - 22
47.4%z 3.63 - 23
47.4%z 3.63 - 24
47.3%z 3.62 - 25
47.2%z 3.61 - 26
47%z 3.60 - 27
46.8%z 3.59 - 28
46.6%z 3.58 - 29
46.3%z 3.55 - 30
45.9%z 3.53 - 31
45.6%z 3.51 - 32
45.4%z 3.50 - 33
44.9%z 3.47 - 34
44.6%z 3.45 - 35
44.5%z 3.44 - 36
44.4%z 3.43 - 37
43.8%z 3.39 - 38
43.4%z 3.37 - 39
43.2%z 3.36 - 40
43.1%z 3.35 - 41
42.9%z 3.34 - 42
42.7%z 3.33 - 43
42.1%z 3.28 - 44
41.8%z 3.27 - 45
41.7%z 3.26 - 46
41.5%z 3.24 - 47
41.5%z 3.24 - 48
41.3%z 3.23 - 49
40.7%z 3.19 - 50
40.5%z 3.18