Marco Fu: 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
95.5%z 5.87 - 2
76.3%z 3.98 - 3
63.8%z 3.84 - 4
60.1%z 3.80 - 5
51.8%z 3.71 - 6
51.1%z 3.70 - 7
51%z 3.70 - 8
50.4%z 3.69 - 9
47.5%z 3.63 - 10
47.4%z 3.63 - 11
46.4%z 3.56 - 12
46.3%z 3.56 - 13
46.1%z 3.54 - 14
46.1%z 3.54 - 15
44.8%z 3.46 - 16
44.6%z 3.45 - 17
44.2%z 3.42 - 18
44%z 3.41 - 19
43.7%z 3.39 - 20
43.7%z 3.39 - 21
43.6%z 3.38 - 22
42.5%z 3.31 - 23
42.4%z 3.30 - 24
42%z 3.28 - 25
41.5%z 3.24 - 26
41.3%z 3.23 - 27
41.2%z 3.22 - 28
40.8%z 3.20 - 29
40.2%z 3.16 - 30
40%z 3.15 - 31
39.7%z 3.13 - 32
39.6%z 3.12 - 33
39.4%z 3.11 - 34
39.3%z 3.11 - 35
39%z 3.09 - 36
38.8%z 3.07 - 37
38.6%z 3.06 - 38
38.6%z 3.06 - 39
38.3%z 3.04 - 40
38.2%z 3.03 - 41
38%z 3.02 - 42
38%z 3.02 - 43
37.8%z 3.01 - 44
37.7%z 3.00 - 45
37.5%z 2.99 - 46
37.4%z 2.98 - 47
37.4%z 2.98 - 48
37.3%z 2.97 - 49
37.1%z 2.96 - 50
37%z 2.95