Marino Defendi: 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
82.6%z 4.23 - 2
81.5%z 4.14 - 3
81.4%z 4.13 - 4
81%z 4.09 - 5
69.7%z 3.90 - 6
63.5%z 3.84 - 7
60.1%z 3.80 - 8
54.9%z 3.74 - 9
46.8%z 3.59 - 10
46.1%z 3.54 - 11
44.5%z 3.44 - 12
43.1%z 3.35 - 13
43%z 3.34 - 14
42.9%z 3.34 - 15
42.1%z 3.28 - 16
41.9%z 3.27 - 17
41.6%z 3.25 - 18
41.1%z 3.22 - 19
40.8%z 3.20 - 20
40.6%z 3.19 - 21
40.4%z 3.17 - 22
40.3%z 3.17 - 23
40.1%z 3.15 - 24
39.9%z 3.14 - 25
39.7%z 3.13 - 26
39.6%z 3.12 - 27
39.2%z 3.10 - 28
38.8%z 3.07 - 29
38.7%z 3.07 - 30
38.7%z 3.06 - 31
38.6%z 3.06 - 32
38.4%z 3.05 - 33
37.9%z 3.02 - 34
37.8%z 3.01 - 35
37.8%z 3.00 - 36
37.4%z 2.98 - 37
37.3%z 2.97 - 38
37%z 2.95 - 39
37%z 2.95 - 40
36.7%z 2.93 - 41
36.5%z 2.92 - 42
36.5%z 2.92 - 43
36.4%z 2.92 - 44
36.4%z 2.92 - 45
36.2%z 2.90 - 46
35.7%z 2.87 - 47
35.5%z 2.86 - 48
35.4%z 2.85 - 49
35.4%z 2.85 - 50
35.3%z 2.84