Pierre Pevel: 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
59.2%z 3.79 - 2
57%z 3.77 - 3
52.8%z 3.72 - 4
47%z 3.60 - 5
46.9%z 3.60 - 6
46.5%z 3.57 - 7
46.2%z 3.55 - 8
46.1%z 3.54 - 9
45.9%z 3.53 - 10
45.4%z 3.50 - 11
45%z 3.47 - 12
44.7%z 3.45 - 13
43.9%z 3.40 - 14
43.4%z 3.37 - 15
42.6%z 3.32 - 16
41.7%z 3.26 - 17
41.7%z 3.26 - 18
41.6%z 3.25 - 19
41.6%z 3.25 - 20
41.2%z 3.23 - 21
41%z 3.21 - 22
40.4%z 3.18 - 23
40.4%z 3.17 - 24
40.3%z 3.17 - 25
39.9%z 3.14 - 26
39.7%z 3.13 - 27
39.7%z 3.13 - 28
39.5%z 3.12 - 29
39.4%z 3.11 - 30
39.1%z 3.09 - 31
39%z 3.08 - 32
38.6%z 3.06 - 33
38.6%z 3.06 - 34
38.4%z 3.04 - 35
38.3%z 3.04 - 36
38.2%z 3.03 - 37
38%z 3.02 - 38
37.7%z 3.00 - 39
37.3%z 2.98 - 40
37.3%z 2.97 - 41
37.2%z 2.97 - 42
37.2%z 2.96 - 43
37.1%z 2.96 - 44
37%z 2.96 - 45
36.8%z 2.94 - 46
36.7%z 2.94 - 47
36.6%z 2.93 - 48
36.5%z 2.92 - 49
36.3%z 2.91 - 50
36.1%z 2.90