Marco Castelli: 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
83.2%z 4.28 - 2
80.4%z 4.05 - 3
80.1%z 4.02 - 4
80%z 4.02 - 5
75.7%z 3.97 - 6
71.3%z 3.92 - 7
71%z 3.92 - 8
68.9%z 3.90 - 9
53.9%z 3.73 - 10
53.8%z 3.73 - 11
53.2%z 3.72 - 12
47.7%z 3.65 - 13
47.4%z 3.63 - 14
46.8%z 3.59 - 15
46.6%z 3.58 - 16
46.6%z 3.58 - 17
46.2%z 3.55 - 18
45.3%z 3.49 - 19
44.9%z 3.47 - 20
44.5%z 3.44 - 21
44.2%z 3.42 - 22
44%z 3.40 - 23
43.4%z 3.37 - 24
43.4%z 3.37 - 25
43.2%z 3.35 - 26
43%z 3.34 - 27
43%z 3.34 - 28
42.9%z 3.34 - 29
42.7%z 3.33 - 30
42.5%z 3.31 - 31
42.2%z 3.29 - 32
42.2%z 3.29 - 33
42.1%z 3.28 - 34
41.9%z 3.27 - 35
41.8%z 3.27 - 36
41.8%z 3.27 - 37
41.6%z 3.25 - 38
41.5%z 3.24 - 39
41%z 3.21 - 40
40.9%z 3.21 - 41
40.9%z 3.20 - 42
40.5%z 3.18 - 43
40.1%z 3.16 - 44
39.8%z 3.14 - 45
39.8%z 3.13 - 46
39.7%z 3.13 - 47
39.3%z 3.11 - 48
39.3%z 3.10 - 49
39.1%z 3.09 - 50
38.9%z 3.07