Marco Lanna: 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
96.8%z 6.12 - 2
93.6%z 5.52 - 3
92.8%z 5.36 - 4
91.2%z 5.05 - 5
90.4%z 4.90 - 6
90.2%z 4.87 - 7
89.8%z 4.82 - 8
88.9%z 4.75 - 9
88.7%z 4.72 - 10
88.4%z 4.70 - 11
87.2%z 4.60 - 12
86%z 4.51 - 13
85.8%z 4.49 - 14
84.3%z 4.37 - 15
82.6%z 4.23 - 16
82.5%z 4.22 - 17
82.2%z 4.20 - 18
81.2%z 4.11 - 19
81.1%z 4.10 - 20
78%z 3.99 - 21
75.9%z 3.97 - 22
73.5%z 3.95 - 23
72.4%z 3.93 - 24
66.8%z 3.87 - 25
61.2%z 3.81 - 26
58.5%z 3.78 - 27
58%z 3.78 - 28
55.7%z 3.75 - 29
55.6%z 3.75 - 30
54.6%z 3.74 - 31
53.7%z 3.73 - 32
53.3%z 3.72 - 33
53%z 3.72 - 34
51.9%z 3.71 - 35
50.9%z 3.70 - 36
50.3%z 3.69 - 37
49.3%z 3.68 - 38
48.9%z 3.68 - 39
48.4%z 3.67 - 40
48%z 3.67 - 41
47.3%z 3.62 - 42
47.1%z 3.61 - 43
47%z 3.60 - 44
46.5%z 3.57 - 45
46%z 3.54 - 46
46%z 3.54 - 47
45.6%z 3.51 - 48
45.3%z 3.49 - 49
45.2%z 3.48 - 50
45%z 3.47