Leo Yang: 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
91.2%z 5.06 - 2
90.3%z 4.89 - 3
90.1%z 4.86 - 4
90.1%z 4.85 - 5
90%z 4.84 - 6
83.1%z 4.27 - 7
82.4%z 4.21 - 8
82.1%z 4.19 - 9
81.3%z 4.12 - 10
80.6%z 4.06 - 11
80.3%z 4.04 - 12
79.7%z 4.01 - 13
78.5%z 4.00 - 14
76.5%z 3.98 - 15
71.9%z 3.93 - 16
69%z 3.90 - 17
66.6%z 3.87 - 18
64.2%z 3.84 - 19
63.3%z 3.83 - 20
59.6%z 3.79 - 21
59.1%z 3.79 - 22
58.2%z 3.78 - 23
57.8%z 3.77 - 24
55.8%z 3.75 - 25
55.6%z 3.75 - 26
53.8%z 3.73 - 27
53.1%z 3.72 - 28
51.2%z 3.70 - 29
51%z 3.70 - 30
49.1%z 3.68 - 31
48%z 3.66 - 32
47.8%z 3.65 - 33
46.7%z 3.58 - 34
46.5%z 3.57 - 35
46.4%z 3.56 - 36
46.3%z 3.55 - 37
46%z 3.54 - 38
45.8%z 3.52 - 39
45.8%z 3.52 - 40
45.6%z 3.51 - 41
45.2%z 3.48 - 42
44.8%z 3.46 - 43
44.7%z 3.46 - 44
44.6%z 3.45 - 45
44.5%z 3.44 - 46
44.5%z 3.44 - 47
44.5%z 3.44 - 48
44.4%z 3.43 - 49
44.4%z 3.43