Matthew Cheung: 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
71.8%z 3.93 - 2
67.1%z 3.88 - 3
64.3%z 3.85 - 4
48.4%z 3.67 - 5
47.9%z 3.66 - 6
46.5%z 3.57 - 7
45.1%z 3.48 - 8
44.9%z 3.46 - 9
43.9%z 3.40 - 10
43.4%z 3.37 - 11
42.6%z 3.32 - 12
41.9%z 3.27 - 13
41.4%z 3.24 - 14
41.4%z 3.24 - 15
40.6%z 3.18 - 16
40.3%z 3.17 - 17
40.1%z 3.16 - 18
40.1%z 3.15 - 19
39.7%z 3.13 - 20
39.7%z 3.13 - 21
39.7%z 3.13 - 22
39%z 3.08 - 23
38.7%z 3.06 - 24
38.7%z 3.06 - 25
38.6%z 3.06 - 26
38.4%z 3.04 - 27
38.4%z 3.04 - 28
38.2%z 3.03 - 29
37.9%z 3.01 - 30
37.8%z 3.01 - 31
37.5%z 2.98 - 32
37.3%z 2.97 - 33
37.2%z 2.97 - 34
36.9%z 2.95 - 35
36.9%z 2.95 - 36
36.8%z 2.94 - 37
36.4%z 2.91 - 38
36.3%z 2.91 - 39
36.3%z 2.91 - 40
36.3%z 2.91 - 41
36.2%z 2.90 - 42
35.7%z 2.87 - 43
35.7%z 2.87 - 44
35.6%z 2.86 - 45
35.5%z 2.86 - 46
35.5%z 2.85 - 47
35.4%z 2.85 - 48
35.1%z 2.83 - 49
35.1%z 2.83 - 50
35%z 2.83