Tom Segura: 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
86.8%z 4.57 - 2
82.6%z 4.23 - 3
71.3%z 3.92 - 4
69.6%z 3.90 - 5
47.2%z 3.61 - 6
46%z 3.54 - 7
45%z 3.47 - 8
44.8%z 3.46 - 9
44.5%z 3.44 - 10
44.2%z 3.42 - 11
43.5%z 3.37 - 12
43.2%z 3.36 - 13
43%z 3.34 - 14
42.7%z 3.32 - 15
41.5%z 3.25 - 16
41.4%z 3.24 - 17
41.1%z 3.22 - 18
40.9%z 3.21 - 19
40.7%z 3.19 - 20
39.9%z 3.14 - 21
39.9%z 3.14 - 22
39.5%z 3.12 - 23
39.2%z 3.10 - 24
39.1%z 3.09 - 25
38%z 3.02 - 26
37.9%z 3.01 - 27
37.8%z 3.01 - 28
37.2%z 2.97 - 29
36.4%z 2.91 - 30
36.3%z 2.91 - 31
35.7%z 2.87 - 32
35.6%z 2.86 - 33
35.4%z 2.85 - 34
35.3%z 2.84 - 35
35%z 2.83 - 36
35%z 2.82 - 37
34.9%z 2.82 - 38
34.9%z 2.82 - 39
34.8%z 2.81 - 40
34.8%z 2.81 - 41
34.8%z 2.81 - 42
34.7%z 2.80 - 43
34.4%z 2.79 - 44
34.3%z 2.78 - 45
34.2%z 2.77 - 46
34.2%z 2.77 - 47
34.1%z 2.77 - 48
34%z 2.76 - 49
33.8%z 2.74 - 50
33.6%z 2.73