Daniel Wu: 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
94.1%z 5.60 - 2
90.3%z 4.89 - 3
88.5%z 4.71 - 4
88.4%z 4.70 - 5
88.1%z 4.67 - 6
87.8%z 4.65 - 7
87.4%z 4.62 - 8
85.3%z 4.45 - 9
85.1%z 4.43 - 10
81.3%z 4.13 - 11
81.3%z 4.12 - 12
81.2%z 4.11 - 13
80.6%z 4.07 - 14
80%z 4.02 - 15
73.5%z 3.95 - 16
67.2%z 3.88 - 17
66.3%z 3.87 - 18
65.2%z 3.85 - 19
64%z 3.84 - 20
61.5%z 3.81 - 21
57.3%z 3.77 - 22
56.4%z 3.76 - 23
55.3%z 3.75 - 24
51.2%z 3.70 - 25
50.9%z 3.70 - 26
47.5%z 3.63 - 27
47.3%z 3.62 - 28
47.3%z 3.62 - 29
47.1%z 3.61 - 30
46.8%z 3.59 - 31
46.8%z 3.59 - 32
46.7%z 3.58 - 33
46.5%z 3.57 - 34
46.5%z 3.57 - 35
46.4%z 3.56 - 36
46.3%z 3.56 - 37
46.1%z 3.54 - 38
45.9%z 3.53 - 39
45.8%z 3.53 - 40
45.7%z 3.52 - 41
45.6%z 3.51 - 42
45.4%z 3.50 - 43
45.4%z 3.50 - 44
44.9%z 3.47 - 45
44.8%z 3.46 - 46
44.6%z 3.45 - 47
44.5%z 3.44 - 48
44.5%z 3.44 - 49
44.4%z 3.43