Marco Ventre: 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
83.5%z 4.30 - 3
78.4%z 4.00 - 4
74%z 3.95 - 5
69.4%z 3.90 - 6
66.5%z 3.87 - 7
65.1%z 3.85 - 8
61.9%z 3.82 - 9
61.7%z 3.82 - 10
61.3%z 3.81 - 11
59%z 3.79 - 12
57.3%z 3.77 - 13
54.1%z 3.73 - 14
52.4%z 3.71 - 15
48%z 3.67 - 16
47.8%z 3.66 - 17
47.7%z 3.65 - 18
47.1%z 3.61 - 19
46.8%z 3.59 - 20
46.1%z 3.54 - 21
45.9%z 3.53 - 22
45.4%z 3.50 - 23
45.4%z 3.50 - 24
45.3%z 3.49 - 25
45.2%z 3.48 - 26
45.2%z 3.48 - 27
45%z 3.47 - 28
44.1%z 3.41 - 29
43.9%z 3.40 - 30
43.9%z 3.40 - 31
43.9%z 3.40 - 32
43.8%z 3.39 - 33
43.6%z 3.38 - 34
43.4%z 3.37 - 35
43.3%z 3.36 - 36
43.3%z 3.36 - 37
43.2%z 3.36 - 38
42.9%z 3.34 - 39
42.9%z 3.33 - 40
42.5%z 3.31 - 41
42.5%z 3.31 - 42
41.6%z 3.25 - 43
41.5%z 3.24 - 44
41.4%z 3.24 - 45
41.4%z 3.24 - 46
41.2%z 3.23 - 47
41.2%z 3.23 - 48
41.2%z 3.23 - 49
41.1%z 3.22 - 50
41%z 3.21