Fernando Trueba: 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
93.1%z 5.42 - 2
87.2%z 4.60 - 3
86.5%z 4.55 - 4
83.1%z 4.27 - 5
81.9%z 4.17 - 6
81.7%z 4.16 - 7
81.1%z 4.10 - 8
69.7%z 3.90 - 9
68.6%z 3.89 - 10
58.1%z 3.78 - 11
54.6%z 3.74 - 12
54.1%z 3.73 - 13
52.1%z 3.71 - 14
50.6%z 3.70 - 15
49.2%z 3.68 - 16
47.7%z 3.65 - 17
47.2%z 3.61 - 18
47.2%z 3.61 - 19
46.9%z 3.60 - 20
46.8%z 3.59 - 21
46.2%z 3.55 - 22
46.2%z 3.55 - 23
46%z 3.53 - 24
45.3%z 3.49 - 25
45%z 3.47 - 26
44.8%z 3.46 - 27
44.1%z 3.42 - 28
43.8%z 3.40 - 29
43.3%z 3.36 - 30
42.9%z 3.33 - 31
42.8%z 3.33 - 32
42.4%z 3.30 - 33
42.1%z 3.28 - 34
42%z 3.28 - 35
41.8%z 3.26 - 36
41.7%z 3.26 - 37
41.6%z 3.25 - 38
41.6%z 3.25 - 39
41.3%z 3.23 - 40
41%z 3.21 - 41
40.9%z 3.21 - 42
40.6%z 3.18 - 43
40.4%z 3.17 - 44
40.3%z 3.17 - 45
39.4%z 3.11 - 46
39.4%z 3.11 - 47
39.2%z 3.10 - 48
39.1%z 3.09 - 49
39.1%z 3.09 - 50
39%z 3.08