Matthew Lam: 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
88.1%z 4.68 - 2
85%z 4.42 - 3
84%z 4.34 - 4
80.8%z 4.08 - 5
80.8%z 4.08 - 6
74.4%z 3.96 - 7
68.7%z 3.89 - 8
68.4%z 3.89 - 9
68.4%z 3.89 - 10
66.3%z 3.87 - 11
57.8%z 3.77 - 12
50.6%z 3.70 - 13
48.2%z 3.67 - 14
47.7%z 3.65 - 15
47.4%z 3.63 - 16
47.4%z 3.63 - 17
46.9%z 3.59 - 18
46.8%z 3.59 - 19
46.6%z 3.58 - 20
46.5%z 3.57 - 21
46.1%z 3.55 - 22
45.8%z 3.52 - 23
45.1%z 3.48 - 24
44.7%z 3.45 - 25
44.5%z 3.44 - 26
44.4%z 3.43 - 27
44.3%z 3.42 - 28
44.1%z 3.41 - 29
43.5%z 3.37 - 30
43.3%z 3.37 - 31
43.3%z 3.36 - 32
42.5%z 3.31 - 33
42.5%z 3.31 - 34
42.5%z 3.31 - 35
42.3%z 3.30 - 36
42.2%z 3.29 - 37
42.2%z 3.29 - 38
42.2%z 3.29 - 39
42%z 3.28 - 40
41.8%z 3.27 - 41
41.5%z 3.25 - 42
41.5%z 3.25 - 43
41.4%z 3.24 - 44
41.3%z 3.23 - 45
41.1%z 3.22 - 46
41%z 3.21 - 47
40.9%z 3.21 - 48
40.7%z 3.19 - 49
40.4%z 3.17 - 50
40.3%z 3.16