Mark Durkan: 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
87.6%z 4.64 - 2
77%z 3.98 - 3
76.8%z 3.98 - 4
68.5%z 3.89 - 5
66.1%z 3.86 - 6
60.6%z 3.80 - 7
57.8%z 3.77 - 8
53%z 3.72 - 9
51.2%z 3.70 - 10
47.5%z 3.63 - 11
46.9%z 3.60 - 12
46.7%z 3.58 - 13
45.2%z 3.48 - 14
44.8%z 3.46 - 15
44.5%z 3.44 - 16
44.4%z 3.44 - 17
44.4%z 3.43 - 18
44.2%z 3.42 - 19
44.2%z 3.42 - 20
43.7%z 3.39 - 21
43.4%z 3.37 - 22
43.2%z 3.35 - 23
42.7%z 3.32 - 24
42.2%z 3.29 - 25
42%z 3.28 - 26
41.6%z 3.25 - 27
41.3%z 3.23 - 28
41%z 3.22 - 29
40.6%z 3.19 - 30
40.5%z 3.18 - 31
40.2%z 3.16 - 32
39.8%z 3.14 - 33
39.8%z 3.13 - 34
39.6%z 3.12 - 35
39.4%z 3.11 - 36
39.2%z 3.10 - 37
39.1%z 3.09 - 38
39%z 3.08 - 39
38.9%z 3.07 - 40
38.5%z 3.05 - 41
38.5%z 3.05 - 42
37.9%z 3.01 - 43
37.9%z 3.01 - 44
37.8%z 3.01 - 45
37.7%z 3.00 - 46
37.5%z 2.99 - 47
37.4%z 2.98 - 48
37.4%z 2.98 - 49
37.3%z 2.97 - 50
37.1%z 2.96