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
64.4%z 3.85 - 4
51%z 3.70 - 5
48.4%z 3.67 - 6
47.9%z 3.66 - 7
46.5%z 3.57 - 8
45.1%z 3.48 - 9
44.9%z 3.46 - 10
44.7%z 3.45 - 11
43.9%z 3.40 - 12
43.7%z 3.39 - 13
43.4%z 3.37 - 14
42.6%z 3.32 - 15
42.1%z 3.28 - 16
41.9%z 3.27 - 17
41.4%z 3.24 - 18
41.4%z 3.24 - 19
40.6%z 3.18 - 20
40.3%z 3.17 - 21
40.1%z 3.16 - 22
40.1%z 3.15 - 23
39.7%z 3.13 - 24
39.7%z 3.13 - 25
39.7%z 3.13 - 26
39%z 3.08 - 27
38.7%z 3.06 - 28
38.7%z 3.06 - 29
38.6%z 3.06 - 30
38.4%z 3.04 - 31
38.4%z 3.04 - 32
38.2%z 3.03 - 33
37.9%z 3.01 - 34
37.8%z 3.01 - 35
37.5%z 2.98 - 36
37.3%z 2.97 - 37
37.2%z 2.97 - 38
36.9%z 2.95 - 39
36.9%z 2.94 - 40
36.8%z 2.94 - 41
36.4%z 2.91 - 42
36.3%z 2.91 - 43
36.3%z 2.91 - 44
36.3%z 2.91 - 45
36.2%z 2.90 - 46
36.1%z 2.90 - 47
35.7%z 2.87 - 48
35.7%z 2.87 - 49
35.6%z 2.86 - 50
35.5%z 2.86