Thomas Peter Friedl: 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
88.8%z 4.74 - 2
82.6%z 4.23 - 3
76.3%z 3.98 - 4
65.9%z 3.86 - 5
57.2%z 3.77 - 6
57.1%z 3.77 - 7
54.7%z 3.74 - 8
53.3%z 3.72 - 9
47.7%z 3.65 - 10
46.8%z 3.59 - 11
46.3%z 3.56 - 12
45.4%z 3.50 - 13
45.4%z 3.49 - 14
45%z 3.47 - 15
44.8%z 3.46 - 16
44.7%z 3.45 - 17
44.3%z 3.42 - 18
43.1%z 3.35 - 19
42.5%z 3.31 - 20
41.5%z 3.24 - 21
40.9%z 3.21 - 22
40.8%z 3.20 - 23
40.8%z 3.20 - 24
40.7%z 3.19 - 25
40.6%z 3.19 - 26
40.4%z 3.17 - 27
40.2%z 3.16 - 28
40.1%z 3.16 - 29
39.7%z 3.13 - 30
39.6%z 3.12 - 31
39.3%z 3.10 - 32
39.1%z 3.09 - 33
39%z 3.09 - 34
39%z 3.08 - 35
38.9%z 3.08 - 36
38.5%z 3.05 - 37
38.4%z 3.04 - 38
38.3%z 3.04 - 39
38.3%z 3.04 - 40
38.1%z 3.03 - 41
37.9%z 3.01 - 42
37.8%z 3.01 - 43
37.5%z 2.99 - 44
37.5%z 2.99 - 45
37.2%z 2.97 - 46
37.2%z 2.97 - 47
37%z 2.95 - 48
36.8%z 2.94 - 49
36.5%z 2.92 - 50
36.5%z 2.92