Olivier Kapo: 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
85.2%z 4.44 - 2
62.7%z 3.83 - 3
50.4%z 3.69 - 4
46.3%z 3.56 - 5
45.7%z 3.52 - 6
44.6%z 3.45 - 7
43.6%z 3.38 - 8
43.5%z 3.38 - 9
43.3%z 3.36 - 10
43%z 3.34 - 11
42.9%z 3.33 - 12
41.9%z 3.27 - 13
40.9%z 3.20 - 14
40.8%z 3.20 - 15
40.6%z 3.19 - 16
40.3%z 3.16 - 17
39.9%z 3.14 - 18
39.8%z 3.14 - 19
39.2%z 3.10 - 20
39%z 3.08 - 21
38.7%z 3.06 - 22
38.1%z 3.03 - 23
38.1%z 3.03 - 24
38.1%z 3.02 - 25
38.1%z 3.02 - 26
38%z 3.02 - 27
37.2%z 2.97 - 28
37.2%z 2.97 - 29
36.9%z 2.94 - 30
36.8%z 2.94 - 31
36.7%z 2.94 - 32
36.4%z 2.91 - 33
36.3%z 2.91 - 34
36.2%z 2.90 - 35
36.2%z 2.90 - 36
36%z 2.89 - 37
35.9%z 2.88 - 38
35.9%z 2.88 - 39
35.8%z 2.88 - 40
35.7%z 2.87 - 41
35.6%z 2.87 - 42
35.6%z 2.86 - 43
35.5%z 2.85 - 44
35.4%z 2.85 - 45
35.3%z 2.85 - 46
35.3%z 2.84 - 47
35.3%z 2.84 - 48
35.2%z 2.83 - 49
35%z 2.82 - 50
34.7%z 2.81