Tiago Splitter: 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
95.2%z 5.81 - 2
89.7%z 4.81 - 3
88.7%z 4.73 - 4
85.2%z 4.44 - 5
83.8%z 4.33 - 6
82.9%z 4.25 - 7
82.9%z 4.25 - 8
82.7%z 4.24 - 9
82.6%z 4.23 - 10
82.4%z 4.21 - 11
82%z 4.18 - 12
81.2%z 4.11 - 13
80.8%z 4.08 - 14
80.6%z 4.06 - 15
80.4%z 4.05 - 16
80.4%z 4.05 - 17
80.1%z 4.03 - 18
80.1%z 4.02 - 19
79%z 4.01 - 20
70.3%z 3.91 - 21
69.9%z 3.91 - 22
69.3%z 3.90 - 23
68.4%z 3.89 - 24
67.2%z 3.88 - 25
65.7%z 3.86 - 26
62.9%z 3.83 - 27
62.9%z 3.83 - 28
62.8%z 3.83 - 29
61.3%z 3.81 - 30
58.1%z 3.78 - 31
57.8%z 3.77 - 32
57.5%z 3.77 - 33
50.1%z 3.69 - 34
48.5%z 3.67 - 35
48.1%z 3.67 - 36
48%z 3.66 - 37
47.6%z 3.64 - 38
47%z 3.60 - 39
46.7%z 3.58 - 40
46.5%z 3.57 - 41
46.4%z 3.56 - 42
46.1%z 3.54 - 43
46%z 3.54 - 44
45.9%z 3.53 - 45
45.8%z 3.52 - 46
45.7%z 3.52 - 47
45.7%z 3.52 - 48
45.6%z 3.51 - 49
45.6%z 3.51 - 50
45.5%z 3.51