Marc Forster: 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
85.6%z 4.48 - 2
82.6%z 4.23 - 3
82.2%z 4.19 - 4
81%z 4.10 - 5
76.7%z 3.98 - 6
73.4%z 3.94 - 7
65.8%z 3.86 - 8
64.8%z 3.85 - 9
57.8%z 3.77 - 10
55.7%z 3.75 - 11
47.4%z 3.63 - 12
47.2%z 3.61 - 13
47.1%z 3.61 - 14
46.7%z 3.58 - 15
46.4%z 3.56 - 16
46.3%z 3.56 - 17
46.1%z 3.54 - 18
46.1%z 3.54 - 19
45.8%z 3.52 - 20
45.7%z 3.52 - 21
45.2%z 3.48 - 22
44.4%z 3.43 - 23
44.3%z 3.42 - 24
44%z 3.41 - 25
43.6%z 3.38 - 26
43.5%z 3.38 - 27
43.4%z 3.37 - 28
43%z 3.34 - 29
42.8%z 3.33 - 30
42.7%z 3.32 - 31
42.7%z 3.32 - 32
42.2%z 3.29 - 33
41.8%z 3.26 - 34
41.8%z 3.26 - 35
41.7%z 3.26 - 36
41.5%z 3.25 - 37
41.3%z 3.23 - 38
41.3%z 3.23 - 39
41.3%z 3.23 - 40
41.3%z 3.23 - 41
41.2%z 3.23 - 42
41.2%z 3.23 - 43
41%z 3.21 - 44
41%z 3.21 - 45
40.7%z 3.19 - 46
40.7%z 3.19 - 47
40.5%z 3.18 - 48
40.3%z 3.17 - 49
39.9%z 3.14