Marcos Paquetá: 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
86.3%z 4.53 - 2
75.3%z 3.97 - 3
56.1%z 3.76 - 4
47.8%z 3.65 - 5
45.9%z 3.53 - 6
44.2%z 3.42 - 7
43.8%z 3.40 - 8
43.3%z 3.37 - 9
42.7%z 3.32 - 10
42.6%z 3.31 - 11
41.9%z 3.27 - 12
41.6%z 3.25 - 13
41.1%z 3.22 - 14
41%z 3.22 - 15
40.8%z 3.20 - 16
40.7%z 3.20 - 17
40.4%z 3.17 - 18
39.7%z 3.13 - 19
39.2%z 3.10 - 20
39.2%z 3.09 - 21
38.7%z 3.06 - 22
38.4%z 3.05 - 23
38.4%z 3.04 - 24
38.4%z 3.04 - 25
37.7%z 3.00 - 26
37.7%z 3.00 - 27
37.6%z 2.99 - 28
37.3%z 2.97 - 29
37.2%z 2.97 - 30
37.1%z 2.96 - 31
36.5%z 2.92 - 32
36.5%z 2.92 - 33
36.5%z 2.92 - 34
36.3%z 2.91 - 35
36.3%z 2.91 - 36
36.3%z 2.91 - 37
36.2%z 2.90 - 38
36.1%z 2.90 - 39
35.9%z 2.88 - 40
35.2%z 2.83 - 41
35.1%z 2.83 - 42
34.8%z 2.81 - 43
34.8%z 2.81 - 44
34.7%z 2.80 - 45
34.6%z 2.80 - 46
34.6%z 2.80 - 47
34.4%z 2.78 - 48
34.2%z 2.77 - 49
34.1%z 2.77 - 50
34.1%z 2.76