Tomo Hafner: 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
45.4%z 3.50 - 2
44.2%z 3.42 - 3
42.9%z 3.34 - 4
41.7%z 3.26 - 5
41.6%z 3.25 - 6
40.8%z 3.20 - 7
40.8%z 3.20 - 8
39.8%z 3.14 - 9
39.6%z 3.12 - 10
39.6%z 3.12 - 11
39.5%z 3.12 - 12
39.4%z 3.11 - 13
39%z 3.08 - 14
38.7%z 3.06 - 15
38.5%z 3.05 - 16
38.4%z 3.05 - 17
38.4%z 3.05 - 18
38.1%z 3.03 - 19
37.9%z 3.01 - 20
37.6%z 2.99 - 21
37.4%z 2.98 - 22
36.9%z 2.95 - 23
36.9%z 2.95 - 24
36.2%z 2.90 - 25
35.9%z 2.88 - 26
35.4%z 2.85 - 27
35.2%z 2.84 - 28
34.7%z 2.80 - 29
34.5%z 2.79 - 30
34.5%z 2.79 - 31
34.4%z 2.79 - 32
34.3%z 2.78 - 33
34.2%z 2.77 - 34
34.1%z 2.76 - 35
33.9%z 2.75 - 36
33.8%z 2.74 - 37
33.6%z 2.74 - 38
33.6%z 2.73 - 39
33.5%z 2.73 - 40
33.3%z 2.72 - 41
33.3%z 2.71 - 42
33.2%z 2.71 - 43
32.9%z 2.69 - 44
32.8%z 2.68 - 45
32.8%z 2.68 - 46
32.8%z 2.68 - 47
32.7%z 2.68 - 48
32.6%z 2.67 - 49
32.6%z 2.67 - 50
32.4%z 2.66