Marco Ferrante: 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
45.5%z 3.50 - 2
43.7%z 3.39 - 3
43.4%z 3.37 - 4
41.9%z 3.27 - 5
41.5%z 3.24 - 6
41%z 3.21 - 7
40.2%z 3.16 - 8
39.6%z 3.12 - 9
39.4%z 3.11 - 10
39.3%z 3.10 - 11
39.3%z 3.10 - 12
39.2%z 3.10 - 13
38.8%z 3.07 - 14
38.7%z 3.06 - 15
38.3%z 3.04 - 16
38.1%z 3.02 - 17
37.8%z 3.01 - 18
37.8%z 3.01 - 19
37.8%z 3.01 - 20
37.6%z 2.99 - 21
37.4%z 2.98 - 22
37.2%z 2.97 - 23
37%z 2.95 - 24
36.7%z 2.93 - 25
36.6%z 2.93 - 26
36.5%z 2.92 - 27
36.4%z 2.92 - 28
35.6%z 2.86 - 29
35.5%z 2.85 - 30
35.2%z 2.84 - 31
35.2%z 2.83 - 32
35.1%z 2.83 - 33
34.5%z 2.79 - 34
34.5%z 2.79 - 35
34.4%z 2.78 - 36
34.4%z 2.78 - 37
34.3%z 2.78 - 38
34.3%z 2.78 - 39
34.3%z 2.78 - 40
34.3%z 2.78 - 41
34.2%z 2.77 - 42
33.7%z 2.74 - 43
33.7%z 2.74 - 44
33.6%z 2.74 - 45
33.6%z 2.73 - 46
33.6%z 2.73 - 47
33.3%z 2.71 - 48
33.2%z 2.71 - 49
33.1%z 2.70 - 50
33%z 2.70